SAYINGS OF CONFUCIUS, DESELECTED
Volume 1
Michael Justin Hunter
A DISSERTATION
PRESENTED TO THE FACULTY
OF PRINCETON UNIVERSITY
IN CANDIDACY FOR THE DEGREE
OF DOCTOR OF PHILOSOPHY
RECOMMENDED FOR ACCEPTANCE
BY THE DEPARTMENT OF
EAST ASIAN STUDIES
Adviser: Willard Peterson
September 2012
© Copyright by Michael Justin Hunter, 2012. All rights reserved.
Abstract
In this dissertation I set out to question the status of the Analects (Lunyu 論語, or The Selected
Sayings) as the most authoritative source of Confucius’s teachings and, in the process, to rethink
Confucius’s place in early Chinese intellectual history.
Part one is a roadmap to the thousands of Confucius sayings preserved in the early corpus. It
includes an in-depth discussion of Confucius quotation markers as well as a comprehensive list
of texts that quote Confucius. In the course of the chapter, I argue that Confucius quotation before the advent of the Analects was a dynamic, creative practice in which authors treated Confucius sayings as venues for the re-performance of inherited wisdom.
Part two presents the main argument for revising the dating of the Analects based on a reverse chronological survey of Confucius quotation practice in the early period. I conclude the
chapter with the argument that the Analects was compiled between the 150s and 130s BCE,
roughly three centuries later than the traditional account would have it.
Part three brings the focus back to the Analects itself to read the text as a product of a Western Han political, intellectual, and textual milieu. The chapter develops as a series of nine case
studies, each of which identifies a different compilation strategy employed by the Analects
compilers.
In a concluding chapter, I synthesize the arguments of the first three parts and discuss the implications of my findings for the study of early Chinese thought. In an epilogue, I also develop
one concrete scenario for the creation of a Western Han Analects.
i
For Catherine
ii
TABLE OF CONTENTS
– Volume 1 –
◇ INTRODUCTION.........................................................................................................1
□ The Kongzi problem ...............................................................................................1
□ The traditional view of the Lunyu ...........................................................................4
□ Against the traditional view ....................................................................................7
□ A tradition of critical Kongzi/Lunyu scholarship..................................................10
◇ CHAPTER ONE: AN INTRODUCTION TO CONFUCIUS SAYINGS .................24
□ What is a Kongzi saying?......................................................................................25
□ How are Kongzi sayings identified? .....................................................................32
□ Where are Kongzi sayings found? ........................................................................38
□ How are Kongzi sayings used in early texts?........................................................43
○ Dependent Kongzi sayings...........................................................................47
– Kongzi comments................................................................................47
• Comments on anecdotes..............................................................48
• Comments on other sayings ........................................................53
• Comments on people...................................................................59
• Comments on natural phenomena...............................................67
– Kongzi commentaries..........................................................................68
• Amplifying commentaries...........................................................71
• “Greater meaning” (da yi 大義) commentaries ..........................74
• “Clarifying” (ming 明) commentaries.........................................78
• Word and character glosses.........................................................81
• Meta-commentaries.....................................................................82
○ Independent Kongzi sayings ........................................................................83
– Objects of commentary........................................................................87
– Objects of criticism ............................................................................94
– Proof texts ..........................................................................................98
– Records of Kongzi’s life and thought................................................101
○ “Imputed words” (yu yan 寓言).................................................................104
□ How did early authors conceive of Kongzi quotation practice? .........................115
□ Conclusions.........................................................................................................120
iii
◇ CHAPTER TWO: A BRIEF HISTORY OF CONFUCIUS QUOTATION PRACTICE,
FROM 220 CE THROUGH THE FOURTH CENTURY BCE
.....................133
□ Kongzi quotation practice in the early period: The Big Picture..........................134
□ The Eastern Han period (25–220 CE).................................................................143
○ Edicts..........................................................................................................143
○ Commentaries ............................................................................................148
○ Master (zi 子) texts.....................................................................................153
○ Bibliography...............................................................................................156
□ The Xin period (9–23 CE) ..................................................................................158
□ The late Western Han period (100 BCE–9 CE) ..................................................162
○ Yang Xiong 揚雄 (53 BCE–18 CE)...........................................................162
○ Liu Xiang 劉向 (79–8 BCE)......................................................................164
○ Edicts..........................................................................................................167
○ Memorials ..................................................................................................173
○ Histories .....................................................................................................181
□ Interlude: Searching for signs of the Lunyu ........................................................187
□ The early Western Han period (203–100 BCE)..................................................189
□ The Pre-Han period .............................................................................................196
○ The easy cases ............................................................................................196
○ The hard cases ............................................................................................199
□ Lunyu parallels outside the corpus of Kongzi sayings ........................................206
□ Conclusion: The Lunyu as Western Han text......................................................207
◇ CHAPTER THREE: HOW THE LUNYU WAS LUN-ED: NINE THEORIES
ON THE COMPILATION OF THE ANALECTS ...............................................213
□ Lunyu 13/2 and “promoting the worthy and talented” ........................................217
□ Keeping up with the Laozi...................................................................................224
□ Lunyu 2/19 & 14/21 and white-washed Kongzi sayings.....................................231
□ Epitomizing Kongzi sayings ...............................................................................235
□ Kongzi on Kongzi ...............................................................................................241
□ Lunyu 12/1 & 12/2 and the situated saying.........................................................245
□ Lunyu 7/1 and the appropriated Kongzi saying ..................................................249
□ Template sayings.................................................................................................255
□ A Zengzi layer? ...................................................................................................261
iv
□ Conclusion ..........................................................................................................265
◇
CONCLUSION: THE PLACE OF CONFUCIUS AND THE ANALECTS
IN EARLY CHINESE THOUGHT ....................................................................270
◇
AN IMAGINATIVE EPILOGUE: THE COMPILATION AND RISE
OF THE ANALECTS, CIRCA 150 BCE–82 BCE ..............................................285
– Volume 2 –
◇ CHAPTER ONE APPENDICES ..............................................................................298
□ Appendix A: “Between Chen and Cai” stories through the Western Han..........298
□ Appendix B: Passages in which Kongzi enjoins disciples to
“take note of” (ji zhi 記之/zhi zhi 志之/shi zhi 識之) or
“work on” (mian zhi 勉之) his words............................................305
□ Appendix C: The Shanghai Museum “Lu bang da han” manuscript ..................303
□ Appendix D: Zigao debunks Kongzi rumors in the Kong congzi .......................305
□ Appendix E: Mengzi debunks Kongzi rumors at Mengzi 5A/8 ..........................307
◇ CHAPTER TWO APPENDICES..............................................................................308
□ Appendix A: Lunyu 13/18 parallels ....................................................................308
□ Appendix B: Kongzi attributions in Eastern Han edicts .....................................309
□ Appendix C: Kongzi attributions in Eastern Han commentaries ........................310
□ Appendix D: The Lunyu entry in the “Yiwen zhi”..............................................311
□ Appendix E: Lunyu transmitters mentioned in the “Yiwen zhi”.........................313
□ Appendix F: Wang Chong’s history of the Lunyu ..............................................314
□ Appendix G: A Shuiyuan Kongzi attribution......................................................316
□ Appendix H: Kongzi attributions in Western Han edicts ...................................317
□ Appendix I: Emperor Wu’s 123 BCE Kongzi attribution and its parallels.........318
□ Appendix J: Kongzi attributions in Han memorials............................................319
□ Appendix K: Lunyu Kongzi sayings in the Hanshu organized by Lunyu book ..320
□ Appendix L: Lunyu Kongzi sayings in the Shiji organized by Lunyu book........322
□ Appendix M: Lunyu parallels in Huainanzi Kongzi sayings ..............................324
□ Appendix N: Lunyu parallels in Lüshi chunqiu Kongzi sayings.........................326
v
□ Appendix O: Lunyu parallels in Gongyang Kongzi sayings...............................330
□ Appendix P: Lunyu parallels in Han Feizi Kongzi sayings ................................331
□ Appendix Q: Lunyu parallels in Xunzi Kongzi sayings ......................................332
□ Appendix R: Lunyu parallels in Zhuangzi Kongzi sayings.................................334
□ Appendix S: Lunyu parallels in Mozi Kongzi sayings ........................................337
□ Appendix T: Lunyu parallels in Zuozhuan Kongzi sayings ................................339
□ Appendix U: Lunyu 13/16 parallels in early texts...............................................342
□ Appendix V: Lunyu parallels in Shanghai Museum corpus Kongzi sayings ......343
□ Appendix W: Lunyu parallels in Mengzi Kongzi sayings...................................347
□ Appendix X: Lunyu parallels outside the corpus of Kongzi sayings ..................356
◇ CHAPTER THREE APPENDICES...........................................................................387
□ Appendix A: Emperor Wen’s 165 BCE edict .....................................................387
□ Appendix B: Lunyu 5/26 & 11/26.......................................................................389
□ Appendix C: Lunyu 5/26 & 11/26 parallels ........................................................391
□ Appendix D: Lunyu 2/17 and its parallels...........................................................392
□ Appendix E: Second-order knowledge in the Laozi, Lüshi chunqiu,
Zhuangzi, Huainanzi, and Wenzi ...................................................394
□ Appendix F: Lunyu 14/34 parallels in Western Han sources..............................396
□ Appendix G: Lunyu 9/26 parallels ......................................................................397
□ Appendix H: Lunyu 14/35 parallels ....................................................................399
□ Appendix I: Lunyu 14/21 and the Zuozhuan.......................................................401
□ Appendix J: Lunyu 2/19 and the Huainanzi........................................................402
□ Appendix K: Lunyu 9/28 parallels ......................................................................403
□ Appendix L: Lunyu 9/17 parallels.......................................................................404
□ Appendix M: Lunyu 9/17 and the Chuci.............................................................405
□ Appendix N: Lunyu 12/1 and the Zuozhuan........................................................406
□ Appendix O: Lunyu 12/2 and the Zuozhuan........................................................407
□ Appendix P: Lunyu 7/1 parallels.........................................................................408
□ Appendix Q: “Erring” (guo 過) and “reforming” (gai 改)
in the Lunyu and elsewhere ...........................................................411
□ Appendix R: The junzi 君子/xiaoren 小人 dichotomy in the Lunyu..................413
□ Appendix S: The “states with the Way”/“states without the Way”
dichotomy in the Lunyu .................................................................414
□ Appendix T: The “states with the Way”/ “states without the Way”
dichotomy outside the Lunyu........................................................416
vi
□ Appendix U: “Wise men”/“humane men”/“brave men” comparisons
in the Lunyu ...................................................................................418
□ Appendix V: “Wise men”/“humane men”/“brave men” comparisons
outside the Lunyu...........................................................................419
□ Appendix W: Lunyu Zengzi material..................................................................421
□ Appendix X: Liu De’s 劉德 Hanshu biography and related passages................422
◇ BIBLIOGRAPHY .....................................................................................................424
□ Primary sources...................................................................................................424
□ Secondary sources and translations.....................................................................426
vii
Acknowledgements
This dissertation would not have been possible without the help and encouragement of various
professors, friends, and family, beginning with my advisers. Willard Peterson and Martin Kern
have been unflagging in their support and enthusiasm for this project. My approach to the study
of early Chinese thought and textual culture owes everything to their influence and training.
Robert Bagley, Benjamin Elman, Andrew Ford, Susan Naquin, Andrew Plaks, and Wang Ping
also provided valuable feedback at various stages of this project’s genesis.
I would also like to thank the participants in the “Analects: A Western Han Text?” conference held at Princeton University on November 4–5, 2011, for their feedback on some of the arguments which made their way into this dissertation: Wolfgang Behr, Anne Cheng, Mark Csikszentmihalyi, Paul van Els, Joachim Gentz, Paul R. Goldin, Martin Kern, Esther Klein, John
Makeham, Cameron Moore, Matthias Richter, David Schaberg, Wojciech Simson, Ken’ichi
Takashima, and Oliver Weingarten. Ongoing conversations with Oliver Weingarten in particular
have benefited this project in any number of ways.
My final year of study was supported by a dissertation completion fellowship from the
Mellon/American Council of Learned Societies Early Career Fellowship Program.
Finally, I would like to thank my father, Jim, and my sister, Meaghan, for their encouragement; Kieran and Susanna for their advice on all things academic; James and Kieran for being
such adorable distractions; and, above all, my wife Catherine for her patience and steadfast
support.
viii
INTRODUCTION
The Kongzi Problem1
Open any introduction to Chinese civilization and you will find a section toward the beginning
about Kongzi 孔子 (551–479 BCE), the figure known to Western audiences as Confucius.2 You
learn that Kongzi lived from the mid-sixth through the early fifth centuries BCE; that he was
from the state of Lu 魯, one of the smaller of the “Warring States” (Zhanguo 戰國) struggling to
defend itself against its more powerful neighbors in a time of constant warfare; that, try as he
might, Confucius failed to find employment as an advisor to a lord; and that he taught a number
of disciples who transmitted his teachings to later generations. Next comes philosophy: one is
told that Kongzi was the first Chinese philosopher;3 that he sought to revivify the ideals of the
Zhou 周 sage-kings; and that he upheld li 禮 (“ritual”) and ren 仁 (“benevolence”) as the Dao 道
(Way) to bring order to a chaotic age. Kongzi is also said to have founded an -ism—“Confucianism”—that would become a cornerstone of East Asian culture, and he is credited with having
created, edited, or otherwise inspired the “Confucian” classics, the foundational curriculum of
elite Chinese education up until the twentieth century. Finally, you learn a methodology for
studying Kongzi’s thought: consult the Lunyu 論語 (Analects, or Selected Sayings), the oldest
and most authoritative source of Kongzi’s teachings.4
1. This heading is inspired by Louis-André Dorion’s 2011 essay, “The Rise and Fall of the Socratic Problem.” See
my conclusion, pp. 275–277, for additional discussion of the parallels between Socrates and Kongzi.
2.
See, e.g., Ebrey 1996, pp. 42–43.
3.
See, e.g., Nivison 1999, pp. 746 & 754.
4. See, e.g., Lau 2000, p. ix. An excellent illustration of the Lunyu’s dominance of Kongzi studies is Luo Anxian’s
羅安憲 Zhongguo Kong xue shi 中國孔學史 (A History of Kongzi Studies in China), which devotes a mere thirteen
pages to the question of “Kongzi sixiang shiliao 孔子思想史料” (“Sources of Kongzi’s Thought”), eight of which
1
Complications arise when we ask the necessary question: how do we know what we think we
know about Kongzi? What are our sources for the historical Kongzi, and how reliable are they?5
Even a cursory review of the relevant texts reveals huge gaps in the historical record. The earliest
extant biography of Kongzi, Sima Qian’s 司馬遷 (d. ca. 86 BCE) “Kongzi shijia 孔子世家”
(“Hereditary House of Kongzi”), is dated to the middle part of the Western Han 西漢 dynasty
(202 BCE–9 CE), roughly three and a half centuries after Kongzi is supposed to have died.6 The
Chunqiu 春秋 (Springs and Autumns) chronicle, the text which anchors Kongzi’s traditional
chronology, only mentions Kongzi in a single, three-character biographical tidbit in its entry for
479 BCE: “Kong Qiu passed away” (Kong Qiu zu 孔丘卒).7 While the Zuozhuan 左傳 (Zuo Traditions) commentary to the Chunqiu fills in a few more biographical details, the dating of this
text remains controversial.8 Nor is the Lunyu itself a self-evidently ancient or authentic source of
Kongzi’s teachings. The earliest extant source to describe the Lunyu as a fifth-century text dates
no earlier than the late Western Han period, more than four centuries after the fact.9 Still other
sources of Kongzi biography and doxography like the Mengzi 孟子 (Mencius) are of uncertain
dating and/or were composed centuries after Kongzi’s death.10
deal with the Lunyu. Zhu Weizheng (2002, p. 98) has observed that only a handful of the hundreds of scholarly
articles on Kongzi’s thought published between 1949 and the 1980s make use of non-Lunyu Kongzi material.
5. See especially Zhu Weizheng 1987 for a strong statement of the argument that one cannot discuss the history of
Kongzi without also examining the problem of the sources of Kongzi’s thought.
6. Shiji 47.1905–1948. For a translation, see Yang & Yang 1979, pp. 1–27. See also Wu Zhenxun 2009, pp. 227–
28, for an appraisal of the place of the “Kongzi shijia” in Kongzi and Lunyu studies.
7.
Ai 16 (SBCK 30/6b).
8. For this debate, see Pines 2002 and Schaberg 1999. Whereas Pines believes that the text accurately portrays
Spring and Autumn-era realities, Schaberg places the text in the fourth century BCE based on parallels with the
Mengzi.
9.
For a discussion of this source, see p. 4 below.
10. On this point, see also Creel 1949, p. 11, and Haupt 2006, pp. 7–10.
2
Uncertain textual chronologies aside, textbook accounts of Kongzi also tend to downplay
what is arguably Kongzi’s greatest legacy, at least as far as early Chinese textual culture is concerned: the literally thousands of Kongzi sayings preserved throughout the early literature.11 The
phrase Kongzi yue 孔子曰 (Kongzi said) and its variants appear in received texts and excavated
manuscripts from a range of genres that includes treatises, commentaries, imperial edicts, memorials, histories, and anecdote collections. The sheer number of extant Kongzi sayings, many of
which have little to nothing in common with the received Lunyu, is itself a formidable challenge
to the textbook Kongzi. Herein lies what I shall refer to as the “Kongzi problem”: given the diversity and incommensurability of extant Kongzi material, how are we to distinguish the historically authentic material from the fictionalized material?12 Without a reliable criterion of authenticity, any and all attempts to reconstruct Kongzi’s thought are doomed to failure.
This is by no means an original insight. The earliest expression of the Kongzi problem is the
opening of book 50 of the Han Feizi 韓非子, “Xian xue 顯學” (“Showing Off Learning”), a text
traditionally dated to the third century BCE:
Those who show off their learning nowadays are the Ru and Mohists. The Ru venerate Kong
Qiu, the Mohists venerate Mo Di. Since the death of Kongzi there have been the Ru of
Zizhang, the Ru of Zisi, the Ru of the Yan clan, the Ru of the Meng clan, the Ru of the Qidiao
clan, the Ru of the Zhongliang clan, the Ru of the Sun clan, and the Ru of the Yuezheng clan.
Since the death of Mozi there have been the Mohists of the Xiangli clan, the Mohists of the
Xiangfu clan, and the Mohists of the Dengling clan. Thus after Kongzi and Mozi the Ru split
into eight factions and the Mohists into three. What each faction included and excluded
contradicted the others’. Nevertheless, they all claimed to be the true [heirs of] Kongzi and
Mozi. Kongzi and Mozi cannot be resurrected, so who is to settle [the question] of learning
nowadays?
世之顯學，儒、墨也。儒之所至，孔丘也。墨之所至，墨翟也。自孔子之死也，有子張
之儒，有子思之儒，有顏氏之儒，有孟氏之儒，有漆雕氏之儒，有仲良氏之儒，有孫氏
之儒，有樂正氏之儒。自墨子之死也，有相里氏之墨，有相夫氏之墨，有鄧陵氏之墨。
11. Haupt 2006, p. 15–16.
12. On this conundrum see also Tsuda 1946, p. 1 & 501ff, Zhu Weizheng 1987, p. 157, Haupt 2006, pp. 7–8,
Weingarten 2010, p. 4, Csikszentmihalyi 2001, p. 242, and Csikszentmihalyi 2002, p. 134.
3
故孔、墨之後，儒分為八，墨離為三，取舍相反、不同，而皆自謂真孔、墨，孔、墨不
可復生，將誰使定世之學乎？13
The “Xian xue” author was in a polemical mode when he composed this critique and thus might
have exaggerated the intellectual factionalism of his day. But his challenge to contemporary Ru
儒 (and Mohists) still stands: how do we adjudicate conflicting claims about the historical
Kongzi? How do we know which ideas, if any, were truly Kongzi’s?
The traditional view of the Lunyu
For the last two millennia, the most common solution to the Kongzi problem has been what I refer to in this dissertation as “the traditional view of the Lunyu.”14 The earliest known expression
of that view is a brief fragment attributed to Liu Xiang 劉向 (79–8 BCE), the official charged by
Emperor Cheng 成帝 (r. 33–7 BCE) in 26 BCE with creating the first catalogue of the Han imperial library:
All twenty sections of the Lu Lunyu are fine sayings recorded by Kongzi’s disciples.
魯論語二十篇皆孔子弟子記諸善言也.15
The account in the “Yiwen zhi 藝文志” (“Monograph on Arts and Letters”), the thirtieth chapter
of Ban Gu’s 班固 (32–92 CE) Hanshu 漢書 (History of the [Former] Han), drew from Liu Xiang’s earlier catalogue while filling in a few more details:
The Lunyu consists of Kongzi’s responses to his disciples and contemporaries as well as
conversations between his disciples and the words they themselves heard from the Master. At
that time, [Kongzi's] followers each had his own record. After the Master died, his followers
gathered [his sayings] together and selectively edited [the collection], thus calling it the
Selected Sayings.
13. Han Feizi book 50 (Sibu congkan [hereafter SBCK] 19/7a).
14. See Csikszentmihalyi 2001 for an excellent overview of the traditional view of Kongzi, which he bases on the
“Kongzi 孔子” entry in the Ci yuan 詞源 dictionary.
15. This fragment appears in the preface to the Lunyu jijie (SBCK 1/1a).
4
論語者，孔子應答弟子時人及弟子相與言而接聞於夫子之語也。當時弟子各有所記。夫
子既卒，門人相與輯而論篹，故謂之論語。16
According to Ban Gu, the Lunyu collection postdated Kongzi’s death and was compiled from individual collections of Kongzi’s actual sayings as transcribed by his disciples. The disciples did
not include all of Kongzi’s recorded sayings but rather “selectively edited” (lun zuan 論篹) that
material, preserving only the “fine sayings” (shan yan 善言). Liu Xiang and Ban Gu thus articulated a handy criterion for identifying authentic Kongzi sayings: a Kongzi saying is authentic if it
appears in the Lunyu.
Here it must be admitted that talk of a single “traditional view” is something of a straw man.
Even in the early period not all proponents of the traditional view agreed about every detail of
the Lunyu’s history, as Wang Chong 王充 (27–100 CE) makes clear in his polemical account of
Eastern Han Lunyu scholarship (Appendix 2:F).17 Defenders of the traditional view have often
disagreed over, e.g., the precise details of the text’s composition and transmission, the identification of individual Lunyu “layers,” and the question of which Kongzi disciple was ultimately responsible for its compilation.18 Without accounting for all possible variations, a fully elaborated
account of “the traditional view” is bound to misrepresent the opinions of large numbers of
scholars both past and present.
Nevertheless, the traditional view as I understand it can be boiled down to a simple claim
16. Hanshu 30.1716. See chapter two, p. 156, for further discussion of this passage and Appendix 2:D for a
complete translation.
17. For example, Wang Chong departs from the “Yiwen zhi” account when he assert that the Lunyu had been “lost”
(wang 亡) at the founding of the Han dynasty (see Appendix 2:F). Zhu Weizheng (1986, p. 41) also notes that Liu
Xiang and Kuang Heng (d. 30/29 BCE) seem to have disagreed about the content of the text, with Liu Xiang saying
that it recorded “sayings” (yan 言) and Kuang Heng “sayings and deeds” (yan xing 言行).
18. Zhao Zhenxin 1961, pp. 11–16, and Makeham 1996, pp. 2–4, contain useful summaries of various theories of
the composition of the Lunyu from the medieval period through to the twentieth century.
5
which lies at the heart of traditional Lunyu scholarship: the Lunyu is our best source for the sayings of Kongzi because only the Lunyu represents Kongzi’s ipsissima verba as transcribed and
edited by his disciples. Scholars at least as far back as the Western Han asserted Kongzi’s involvement in the creation or transmission of the Five Classics, particularly the Chunqiu chronicle, and these texts have at various times and in various contexts trumped the Lunyu as sources of
Kongzi’s teachings.19 But extracting Kongzi’s wisdom from the Chunqiu entailed a convoluted
hermeneutics whereby Kongzi’s weiyan 微言 (subtle words) were decoded from the formulaic
and laconic language of the chronicle; nowhere does the Chunqiu quote Kongzi directly.20 Here
the Han bibliographers’ emphasis on yan 言 (sayings) and yu 語 (conversations) is crucial: when
dealing with the quotable Kongzi, they tell us, there is no source more authoritative than the
Lunyu.
The late Western Han and early Eastern Han view of the Lunyu was only reinforced in subsequent periods of Chinese history as the text came to be included among the Seven Classics (qi
jing 七經), was eventually re-canonized by Zhu Xi 朱熹 (1130–1200) as one of the Four Books
(Si shu 四書), and was among the very first texts translated by Jesuit missionaries for European
audiences.21 To this day, the Lunyu continues to dominate contemporary thinking about Kongzi
and early Chinese philosophy the world over, including among contemporary scholars of early
19. To take but one example, Mengzi 3B/9 (SBCK 6/13a) treats the Chunqiu as the ultimate source of Kongzi’s
teachings: “Those who understand me do so because of the Chunqiu; those who criticize me do so because of the
Chunqiu” (知我者其惟春秋乎。罪我者其惟春秋乎。). On the association of Kongzi with the Chunqiu in the
early period, see Kern 2011.
20. However, the Chunqiu traditions of the Zuozhuan, Gongyang, and Guliang do make occasional use of Kongzi
quotations.
21. Extracts from the Lunyu and the other Four Books (Si shu 四書) appeared in Confucius Sinarum Philosophus
(Confucius, Philosopher of the Chinese), first published in 1687. For a recent account of this book’s genesis, see
Dew 2009, chapter five (“Printing Confucius in Paris”).
6
China.22 Pedagogical tradition has also played a role in perpetuating the Lunyu’s authority.
Speaking from personal experience, the Lunyu is the very first work of Chinese philosophy I ever
studied and the first classical Chinese text I read in the original, a choice encouraged by standard
introductions to early Chinese philosophy and civilization.23 Pedagogical primacy translated into
historical and philosophical primacy as I came to read the language and ideas of other early texts
against the Lunyu. Generations of scholars raised on the Lunyu have likewise read it as one of, if
not the, core text of early Chinese thought.
Against the traditional view
My goal in the following pages is to demonstrate that the traditional view of the Lunyu is a
flawed approach to the study of Kongzi and of early Chinese thought generally. My argument is
both critical and constructive: in addition to showing the paucity of evidence in support of the
traditional view, I develop my own theory of the Lunyu’s origins and model an entirely different
approach to the study of early Kongzi sayings. I pursue this argument from three perspectives:
Chapter one is a roadmap to the thousands of non-Lunyu Kongzi sayings preserved in anecdote collections, commentaries, treatises, histories, imperial edicts, recently excavated manuscripts, and other sources from the early period. It includes an in-depth discussion of Kongzi
quotation markers as well as a comprehensive list of texts that quote Kongzi. The centerpiece of
the chapter is an overview of the diverse uses of Kongzi sayings in the early corpus supplement-
22. For a recent review of the place of the Lunyu in contemporary Kongzi studies in China, see Makeham 2011.
23. The Lunyu is the subject of chapter two in Chan 1963, chapter three in Schwartz 1985, chapter one in Graham
1989, and chapter one in Ivanhoe & Norden (eds.) 2001.
7
ed with a survey of early authors’ reflections on Kongzi quotation practice.
Such a survey is itself a powerful, if indirect, demonstration of the inadequacy of the traditional view. The limitations of the Lunyu-centric approach to Kongzi become particularly glaring in
light of the sheer number, diversity, and interest of non-Lunyu Kongzi sayings. (My own interest
in this project stems from my accidental discovery of Sun Xingyan’s 孫星衍 [1753-1818]
Kongzi jiyu 孔子集語 Collected Sayings of Kongzi], an early nineteenth-century effort to compile early Kongzi material outside of the Lunyu and a handful of other canonical texts.24) In the
course of the chapter I also venture an answer to the question of why early authors were so interested in Kongzi as a source of quotable wisdom. I argue that Kongzi quotation before the advent
of the Lunyu was a dynamic, creative practice in which authors treated Confucius sayings as
venues for the re-performance of inherited wisdom. Not until the Western Han period did authors
regularly use Confucius sayings as independent teachings worthy of study in their own right.
Moreover, hardly any Kongzi quoters in the pre-Han era evinced an interest in quoting a historically authentic Kongzi.
Chapter two introduces the argument for revising the standard dating of the Lunyu based on a
reverse chronological survey of Kongzi quotation practice from 220 CE through the early part of
the Warring States period. An analysis of the relative distribution of Kongzi sayings with Lunyu
parallels versus those without reveals a clear pattern: whereas authors from the late Western Han
and later periods tended to quote the Lunyu when quoting Kongzi, especially in authoritative
24. The edition of the Kongzi jiyu I have used in this project is the Kongzi jiyu jiaobi 孔子集語校補 (Collected
Sayings of Kongzi, collated and supplemented), which supplements Sun Xingyan’s collection with additional
fragments as well as passages from the Zuozhuan, Xiaojing, Zhouyi, the Mawangdui Yi materials, the Dingzhou
“Rujia zhe yan 儒家者言” manuscript, Mengzi, Liji, and the Shiji “Kongzi shijia” and “Zhongni dizi liezhuan”
biographies. A similar resource is Kongzi—Zhou Qin Han Jin wenxianji (Kongzi–Collected literature from the Zhou,
Qin, Han, and Jin), ed. Wu Genliang, Zhang Ronghua, and Jiang Yihua (1990).
8
contexts like imperial edicts, authors in the early Western Han and earlier periods rarely, if ever,
used Kongzi sayings that paralleled those in the received Lunyu. Close readings of select Kongzi
quotations in a wide range of early texts confirm that authors also did not begin to treat Lunyu
Kongzi sayings as authoritative until ca. 100 BCE, beginning with Sima Qian’s (died ca. 86
BCE) Shiji 史記 (Grand Scribe’s Records). The evidence from Lunyu parallels as well as specific Lunyu mentions points to the beginning of Emperor Wu’s reign (r. 141–87) as a likely date of
compilation. This finding is corroborated, ironically enough, by an examination of Han accounts
of the Lunyu’s history, including the “Yiwen zhi,” which likewise point to Emperor Wu’s reign
as a crucial moment in the text’s history.
• Chapter three shifts the focus onto the Lunyu itself to read the text as the product of a Western Han political, intellectual, and textual milieu. The chapter develops as a series of nine case
studies, each of which identifies a different compilation strategy employed by the Lunyu compilers. These include claiming Kongzi’s authorship of popular proverbs traditions, selecting sayings
to serve as authoritative statements on whole sub-genres of Kongzi lore, and using Kongzi sayings to co-opt proverbs more closely associated with other texts, e.g., the Laozi. In this way, the
Lunyu compilers asserted Kongzi’s ownership of early proverb traditions at the same time as other Han scholars were asserting Kongzi’s authorship of the Five Classics. I also note certain resonances between Lunyu Kongzi sayings and Han imperial edicts having to do with the recruitment
of talented officials, parallels which suggest that the text was conceived in part as a handbook of
character evaluation for imperial princes and aspiring officials alike.
In a final chapter, I summarize the findings of the first three chapters and explore their implications for the study of Kongzi and of early Chinese intellectual history generally. In an epilogue, I also develop a speculative yet historically plausible scenario for the compilation of the
9
Lunyu in the Western Han period.
A tradition of critical Kongzi/Lunyu scholarship25
The overlapping subfields of Kongzi and Lunyu studies have produced an ever-expanding ocean
of scholarship. Compiling an exhaustive summary of this scholarship would be an unenviable
task.26 Organized geographically, it would consist primarily of Chinese and Taiwanese Kongzi
and Lunyu scholarship with Japanese and Korean contributions taking up the bulk of non-Sinophone scholarship, followed by European, North American, and other non-East Asian scholarship in a variety of languages. Organized chronologically, it would begin with Han-era bibliographies and Lunyu commentaries and would include works from successive periods of Chinese
history, eventually including the earliest Japanese and Korean works, the first translations into
Western languages, and then adding an increasing number of studies with each passing century
until the explosion of scholarship that attended the rise of the modern academy in the nineteenth
and twentieth centuries.
Another way of organizing an exhaustive Kongzi/Lunyu bibliography might be according to
whether a given author accepted the traditional view of the Lunyu as the most authoritative
source of Kongzi’s teachings. By that standard we would find that the vast bulk of Kongzi/Lunyu
scholarship from its origins up to the present day occupied roughly 99% of our bibliography,
leaving only a short list of authors at the end of the volume who subjected the traditional view to
25. The introduction to Oliver Weingarten’s recent 2010 dissertation contains a parallel review of the secondary
literature on the Analects.
26. For a few recent efforts, see Fu Yunlung 1981, Zhongguo shehui kexue zhexue yanjiusuo 1987, Li Qiqian et al.
1991a & 1991b, Kunio 1992, and Shaheen 2002.
10
critical scrutiny. It is this sliver of skeptical Lunyu scholarship, and especially studies which have
embraced the notion of a Western Han Lunyu, that I outline below.
The roots of the revisionist position might be traced to the observation, recorded as early as
Zhu Xi’s 朱熹 (1130–1200) Si shu zhangju jizhu 四書章句集注 (Collected Chapter-and-Verse
Commentaries on the Four Books), that the “latter ten books [of the Lunyu] contain many omissions and mistakes” (後十篇多闕誤).27 The same observation motivated Japanese scholar Itō
Jinsai’s 伊藤仁齋 (1627–1705) distinction between the “upper” (shang 上) Lunyu of books 1–10
and the “lower” (xia 下) Lunyu of books 11–20, the latter half of which was thought to have been
added subsequent to the text’s original composition.28 Working along parallel lines, Cui Shu 崔
述 (1740–1816) refined the theory to argue that books 16–20 represent an even later addition
within books 11–20.29 The division of the text into upper and lower sections became a mainstay
of Lunyu studies in the twentieth century, with prominent endorsements by H.G. Creel, Arthur
Waley, D.C. Lau, and E. Bruce and A. Taeko Brooks among others, although scholars have often
disagreed about which books belong in which strata.30
As noted above, the traditional view of the Lunyu as it emerges in the “Yiwen zhi” presents a
ready criterion for distinguishing authentic Kongzi sayings from inauthentic sayings: a saying is
27. Sishu zhangju jizhu 8.173. For this citation see also Bruce and Brooks 1998, pp. 201–202. John Makeham
(1996, p. 6, n. 24) credits the Ming 明 dynasty commentator Luo Yuyi 羅喻義 with being the first to explicitly
divide the text into upper and lower sections.
28. See Itō 1970. Itō’s theory was later elaborated by Ogyū Sorai 荻生徂徠 (1666-1728) and his student Dazai
Shundai 太宰春台 (1680-1747).
29. Zhu Si kaoxin lu, juan 4, pp. 28–29. For a brief summary, see Makeham 1996, pp. 9–10.
30. See Creel 1949, Lau 2000, Brooks & Brooks 1997, Waley 1938. Brooks & Brooks (1997, p. vii) write that the
Lunyu “contains only a core of sayings by the historical Confucius, to which have been added layers of attributed
sayings and conversations invented by his successors to update their heritage, and to address the new needs of
changing times.” For critiques of the Brooks’ proposed chronology, see Slingerland 2000, Schaberg 2001, and
Makeham 1999.
11
authentic if it appears in the Lunyu. By rejecting the “Yiwen zhi” criterion as overly simplistic,
the notion of a layered Lunyu raised the bar for Lunyu studies: the most authentic Kongzi sayings
are those which appear only in the earliest layers of the text. The challenge for Lunyu scholars
then became to determine which sections of the text were truly early.31 In our comprehensive
bibliography of Kongzi/Lunyu studies, however, such theories would occupy a middle ground
between traditional scholarship and more skeptical voices insofar as they maintain the assumption that at least some material in the received Lunyu is authentic. The notion of a core Lunyu has
even strengthened the traditional view by justifying the exclusion of precisely those passages that
challenge the perception of the text as a reliable historical record.
The earliest Chinese scholar to have challenged the traditional view directly was Zhao
Zhenxin 趙貞信, who in his 1961 article “Lunyu jiujing shi shei bianzuan de 《論語》究竟是誰
編纂的” (“Who exactly compiled the Lunyu?”) argued that the text was first compiled in the early Han period during the reign of Emperor 文帝 (r. 180–157) or Jing 景帝 (r. 157–141).32 Drawing on a critical tradition of Qing 清 dynasty scholarship exemplified by Zhang Xuecheng 章學
誠 (1738–1801) and Cui Shu 崔述, Zhao argued that textual culture in the Warring States period
did not allow for “private authorship” (siren zhushu 私人著述) of the sort posited by the traditional view. Not until the early imperial period did written texts become the primary mode of textual transmission, and it was at this time that the first Lunyu text was compiled from material selected from earlier oral traditions. Zhao supported this claim with a survey of Kongzi quotations
31. For a summary and critique of these efforts, see Weingarten 2010, pp. 29–56.
32. Zhao points to three sources for this claim: Zhao Qi’s 趙岐 (129–82) statement in the preface to his Mengzi
commentary (SBCK 序/4a) that an official boshi 博士 (erudite) position was first established for the Lunyu during
Emperor Wen’s reign; the statement in Liu Xin’s 劉歆 (46 BCE–23 CE) Hanshu biography (Hanshu 36.1969) that
“a great number of texts had emerged from all over the empire” (天下眾書往往頗出) during Emperor Wen’s reign;
and the list of Lunyu experts in the “Yiwen zhi,” which includes only two experts from Emperor Wu’s reign and
none from previous reigns (see Appendices 2:D & 2:E).
12
and Lunyu parallels in the the Mengzi’s 孟子, arguing that the Mengzi author could not have
drawn his Kongzi sayings from the Lunyu.33 Although Zhao did not spell it out explicitly, the implication of his proposed dating was clear: the Lunyu’s “value as a historical source” (shiliao jiazhi 史料價值) of Kongzi’s life and thought is not as great as is usually supposed. However,
Zhao’s work did not have any discernible impact on mainstream Kongzi/Lunyu studies in China
or elsewhere.34
The most prominent Chinese scholar to argue along similar lines was Zhu Weizheng 朱維錚,
beginning with his 1986 article “Lunyu jieji cuoshuo 《論語》結集脞說” (“Notes on the compilation of the Lunyu”).35 Although apparently unaware of Zhao’s earlier work, like his predecessor
Zhu framed his argument in terms of the Lunyu’s value as a source of Kongzi’s thought. Noting
the apparent absence of Lunyu quotations in Warring States texts (p. 43) and also the testimony
of the Han Feizi 韓非子 quoted above,36 Zhu placed the advent of the Lunyu to the “ancient text
craze” (gushu re 古書熱) in the first part of the Western Han, specifically during the reigns of
Emperors Jing 景帝 and Wu 武帝 (r. 141–87) when the demand for a collection of Kongzi’s
teachings became especially acute. Prior to that point, Kongzi material circulated in a more
piecemeal and haphazard fashion. However, Zhu also admitted that his proposed theory of the
text, like the traditional view, lacked sufficient supporting evidence, and that the “riddle” (mi 謎)
33. I deal with the relationship between the Lunyu and Mengzi in greater depth in chapter two (pp. 201–206).
34. In an apparent effort to justify his own critical stance toward the Lunyu and to establish a critical tradition of
Lunyu scholarship, in 1969 Zhao edited the Lunyu bian 論語辨 (Discriminations on the Lunyu), a selection of
critical writings on the Lunyu chiefly by Cui Shu 崔述 but also by Kang Youwei 康有為 and others. Zhao also
published articles on the origin of the name of the Lunyu (1936) and on Lunyu book 20 (1962).
35. Zhu (1986, p. 42) traces the skeptical tradition of Lunyu scholarship back to Liu Zongyuan’s 柳宗元 (773–819)
argument in his Lunyu bian 論語辯 (Disputations on the Lunyu) that the Lunyu was compiled by Kongzi’s secondor third-generation disciples. See Liu Zongyuan ji 4.110–11.
36. Zhu Weizheng 1986, p. 43.
13
of the Lunyu’s origins could not be settled.37
A single citation of Takeuchi Yoshio 武內義雄 in Zhao Zhenxin’s 1961 article aside, neither
he nor Zhu Weizheng seem to have been aware of the vibrant tradition of critical Rongo 論語
[=Lunyu] scholarship in Japan in the twentieth century,38 the most important figure of which was
undoubtedly Takeuchi Yoshio 武内義雄, whose Rongo no Kenkyū 論語の研究 (Lunyu Studies)
elaborated the theory of a layered Lunyu in a way which strongly emphasized the Han context.39
In his view, previously circulating independent collections of Kongzi material were first combined into an “ancient-text Lunyu” (gu Lunyu 古論語) in the early part of the Western Han. The
earliest stratum of that text was Lunyu books 2–9, the so-called “Hejian 河間 Lunyu” attested by
Wang Chong and first compiled at the court of Liu De 劉德, posthumously known as King Xian
of Hejian 河間獻王 (r. 155–130/129 BCE), but it also included material from the “Qi Lunyu”
(Qi Lunyu 齊論語) and “Lu Lunyu” (Lu Lunyu 魯論語) collections.40 Other material was subsequently added to the collection over the course of the Han. Although Takeuchi preserved the core
assumption of the traditional view insofar as he believed that some Lunyu layers could be ultimately traced back to Zengzi 曾子, traditionally one of Kongzi’s chief disciples and the author of
the Xiaojing 孝經 (Classic of Filial Piety), his emphasis on the Western Han context had a profound influence on subsequent Rongo scholarship.
37. Zhu further explored the methodological challenges of reconstructing the historical Kongzi in a 1987 article.
38. For these Japanese sources I am indebted to Cameron Moore for his paper “Rongo: The State of the Field: A
Review of Japanese Scholarship on the Composition and Transmission of the Lunyu in the Han,” presented at the
conference “The Analects: A Western Han Text?” held at Princeton University on November 4–5, 2011.
39. Moore (2011, p. 4) notes Takeuchi’s debt to Kanō Naoki, his teacher at Kyōto University whose collected notes
on the Lunyu were published in 1977. Kanō’s writings included the observation that a Warring States-era Lunyu did
not command the authority it did in later periods.
40. For Liu De, see pp. 268–269 and pp. 285–296 below. The “Yiwen zhi” (Appendix 2:D) identifies the “Qi
Lunyu” (Qi lun 齊論) and “Lu Lunyu” (Lu lun 魯論) as two different Lunyu recensions in the Western Han.
14
Two major pre-war contributions that further refined and developed Takeuchi’s conclusions
were Watsuji Tetsurō’s 和
哲郎 (1889–1960) Kōshi 孔子 (Confucius) and Tsuda Sōkichi’s 津
田左右吉 (1873–1961) Rongo to Kōshi no shisō 論語と孔子の思想 (The Analects and the
Thought of Confucius).41 Tsuda in particular was the first Japanese scholar to use evidence from
quotation patterns (specifically, those in the Hanshi waizhuan 韓詩外傳) to date the compilation
of the Lunyu to the Western Han, specifically to the reign of Emperor Wen.42 He argued that, prior to that point, the sayings of the Lunyu represented one among many sets of sayings circulating
among competing groups of Ru, and that the boundaries between these smaller corpora and the
larger corpus of Kongzi sayings were extremely fluid. Tsuda was also the first to seriously question the “Yiwen zhi” account of the Lunyu’s transmission in the Han period, arguing that the notion of an “ancient text” Lunyu was a fabrication and that the Qi and Lu versions differed relatively little from one another. His analysis of quotation patterns in the Mengzi and Xunzi also led
him to question the traditional Lunyu→Mengzi→Xunzi chronology by suggesting that the
Mengzi formed part of the source material for a Western Han Lunyu.
By far the most skeptical voice in twentieth century Lunyu/Rongo studies belonged to Kaneto
Mamoru 金戶守, who in a series of articles from 1970 to 1981 articulated the theory that the
Lunyu was first compiled in the latter part of the Western Han period from a range of sources
that included the Shiji 史記 and Han imperial edicts.43 In his view, the Lunyu Kongzi, like the
Shiji Kongzi, was an amalgamation of pre-Han and early Han Kongzi traditions, and the Lunyu
41. Tsuda’s work was published in 1946 but written in the early 1940s. See Moore 2011.
42. Tsuda also bases this dating on Zhao Qi’s 趙岐 (129–82) claim in his late Eastern Han Mengzi commentary
(SBCK 序/4a) that Emperor Wen 文帝 (r. 180–157 BCE) established an “erudite” (boshi 博士) position for the
Lunyu.
43. Kaneto’s main conclusions are conveniently summarized in his 1979 article.
15
text was in flux up until the end of the Western Han period. Like Tsuda, Kaneto also raised serious questions about the “Yiwen zhi” account of the Lunyu’s transmission, including a critique of
the legend of an “ancient text” Lunyu. At the heart of Kaneto’s study was a detailed comparison
between the Lunyu and its Shiji parallels, in the course of which he endeavored to show (unsuccessfully, in my view) that the Lunyu version in every instance postdated the Shiji.
One crucially important detail unknown to all of the scholars listed thus far is the discovery
of a Lunyu manuscript in a Han-era tomb excavated in 1973 near the city of Dingzhou 定州 in
Hebei 河北 province. Although extremely fragmentary, having suffered from a tomb robbery
and a fire sometime in the early period and then a major earthquake in 1976, the published
transcription parallels the received Lunyu closely enough to place the advent of a text closely resembling the received Lunyu to no later than 55 BCE, the date the tomb was probably closed.
The circulation of the Lunyu even in the far reaches of the Western Han empire has been confirmed by the recent discovery of a fragmentary Lunyu manuscript in a tomb outside of Pyongyang, North Korea, tentatively dated to 45 BCE.44
The next landmark contribution to endorse the theory of a Western Han Lunyu was John
Makeham’s 1996 article “The Formation of Lunyu as a Book.” Basing his argument largely on
the evidence from Lunyu mentions and the lack of significant overlap between Mengzi and Lunyu
Kongzi sayings, Makeham concluded that the Lunyu was not treated as a book until after the discovery of the “ancient text” Lunyu ca. 150–140 BCE and the rise of state-sponsored Confucianism during Emperor Wu’s reign. According to Makeham, this “ancient text” Lunyu had original-
44. The best introduction to the Dingzhou 定州 and Lelang 樂浪 Lunyu manuscripts is Van Els 2011. For a
transcription of the Dingzhou manuscript, see Hebeisheng wenwu yanjiusuo 1997. For the Lelang manuscript, see
Lee et al. 2009 (in Korean), 2010 (Japanese), and 2011 (Chinese).
16
ly consisted of several disparate collections of Kongzi material which were not viewed as
particularly authentic or authoritative in the pre-Qin period, when Kongzi material circulated in a
variety of forms. As for the question of the Lunyu’s value as a source for the historical Kongzi,
Makeham concluded that “it is perhaps best not to be too dogmatic, either in the affirmative or
the negative,” although “clearly there is a need to reconsider the status of other early records of
Confucius’ sayings and conversations with his disciples.”45
One of the few Western scholars to take John Makeham’s proposed dating seriously was
Mark Csikszentimihalyi, who in a series of publications not only developed the theory of a Western Han Lunyu but also explored the implications of that theory for the study of pre-Han intellectual history. In a 2002 article on “Confucius and the Analects in the Han,” he noted that the earliest mentions of the Lunyu in the Hanshu as well as the list of Lunyu transmitters in the “Yiwen
zhi,” most of whom served as tutors to imperial princes, points to the Lunyu’s use as a textbook
for princes of the ruling Liu 劉 clan. In a 2001 chapter on “Confucius” he explored the implications of reading the Lunyu as a selection from earlier textual traditions, suggesting that the “principle of selection” was the primary determinant of the content of the text and not some pristine
transmission lineage.46 Csikszentmihalyi’s 2004 book Material Virtue: Ethics and the Body in
Early China was also the first substantive attempt to model an approach to Warring States intellectual history that did not rely on the traditional view of the Analects. Writing that “[m]inimizing the authorship role of Kongzi has the effect of pulling the rug out from under the usual narrative of what is often called the ‘history of thought’ (sixiang shi 思想史) of early China,”
Csikszentmihalyi proposed that early Ru discourse was refracted through several mutually inde-
45. Makeham 1996, p. 24. Makeham revisited many of these conclusions in a 2009 review article.
46. Csikszentmihalyi 2001, p. 259.
17
pendent disciple traditions, and that untangling these traditions was a more productive approach
than attempting to reconstruct an original, pristine Confucianism.
Two recent doctoral dissertations, one in German and one in English, were the first monograph-length studies in the West to challenge the traditional view of the Lunyu in a substantive
way. In her 2006 thesis, “Und der Meister sprach...: Die Darstellung des Konfuzius in Texten der
Zhanguo- und Frühen Han-zeit” (“And the Master Said...: Representations of Confucius in Warring States and Early Han-era Texts”), Christiane Haupt surveyed the evolving representations of
Kongzi in four texts from the Warring States and early Han (the Zuozhuan, Mengzi, Xunzi, and
Hanshi waizhuan), in each case analyzing these texts’ Lunyu parallels to determine the extent of
the Lunyu’s influence. The rarity of such parallels led Haupt to conclude that the the Lunyu did
not become an authoritative source of Kongzi sayings until the Han period; prior to that point,
the Lunyu was only one among many sources of Kongzi material. Noting that the number of
Lunyu parallels in early texts increases over time, Haupt endorsed the theory that the Lunyu grew
into its present form over many years as more and more sayings were added to the collection.47
Although Haupt provides a number of arguments against automatically taking the Lunyu as a
source for the historical Kongzi, she concluded her dissertation on a note of optimism about the
possibility of distinguishing the authentic material from the inauthentic in the Lunyu.
In his 2010 doctoral dissertation, “Textual Representations of a Sage: Studies of Pre-Qin and
Western Han Sources on Confucius (551–479 BCE),” Oliver Weingarten sought to reorient Lunyu studies “not so much on the Confucius figure itself, but rather on compositional features of
texts that relate to it and their underlying editorial strategies.”48 After critiquing some of the most
47. Haupt 2006, pp. 171–178.
48. Weingarten 2010, p. 5.
18
well known efforts to identify individual Lunyu layers by internal criteria, Weingarten went on to
show how close attention to Lunyu parallels, including those in non-canonical Kongzi materials,
might illuminate the formation of individual Lunyu entries and even entire Lunyu books. He also
used Lunyu parallels to highlight the generic or shared nature of Kongzi material in a number of
early texts, including in the Lunyu. Arguing strongly against efforts to recover the historical
Kongzi, he concluded that “the history of early Confucianism needs to be rewritten first and
foremost as a textual and literary history that charts the mostly imaginary territory created in ancient times by anonymous authors and compilers.”49
*
*
*
The present study builds upon this tradition of critical Lunyu scholarship in a few respects. First,
I believe it is the most comprehensive survey of Kongzi sayings and Lunyu parallels to date,
covering more than 3,000 sayings in dozens received and excavated texts from the early period.
The identification of these Kongzi sayings and Lunyu parallels would not have been possible
without a computer database consisting of multiple digital editions of the early Chinese corpus,
which has allowed me to sift through vast amounts of material in ways that would have been impractical, if not impossible, using traditional research tools.50 Using this database, I marked and
collected each and every quotation of Kongzi (or The Master [zi 子]) in the early literature, and
this corpus formed the basis for chapter one’s survey of Kongzi quotation practice in the early
period.51
49. Weingarten 2010, p. 201.
50. The program I used to build this database was BBEdit (Bare Bones Software, http://www.barebones.com/) for
Mac OSX. Although designed primarily for computer programmers and website developers, it offers a number of
tools for organizing and searching through numerous large text files.
51. Most of these Kongzi sayings had already been collected by the modern editors of the Kongzi jiyu jiaobu and
Kongzi—Zhou Qin Han Jin wenxianji compilations. However, assembling this material in digital format allowed me
19
The same database tools also facilitated the study of Lunyu parallels by enabling “fuzzy”
searching, i.e., the identification of parallel passages which are not simply word-for-word matches.52 For instance, instead of searching for the simple search string shu er bu zuo 述而不作
(transmit without originating), the first four characters of the Kongzi saying at Lunyu 7/1, I was
able to search for shu 述 within five (or ten or twenty) spaces of zuo 作, or shu 述 (or its phonetic and semantic variants shu 術 and xun 循) within a certain number of spaces of zuo 作, thereby
yielding each and every passage in which shu 述 is coordinated with zuo 作 irrespective of the
intervening text.53 After running fuzzy searches for each and every phrase in the received Lunyu
across the entire early corpus, I was able to compile a collection of more than one thousand Lunyu parallels from the earliest Warring States texts through the end of the Eastern Han period.
This data facilitated the study of Lunyu parallels in chapters two and three.54
The second contribution of this dissertation is conceptual. In contrast to most of the studies
listed above, this dissertation does not take Kongzi as its primary object of analysis.55 Instead, I
am concerned with “Kongzi quotation practice,” my term for the loose set of conventions which
dictated what kinds of material could be attributed to Kongzi in which contexts. One of the advantages of this approach is that it maintains the focus on what was arguably Kongzi’s greatest
appeal to early authors: his quotability. While Kongzi the man appears as a character and as a
to run more targeted searches within the corpus of early Kongzi material.
52. This is the “grep” (“Global/Regular Expression/Print) search utility. For a general introduction, see http:/
/en.wikipedia.org/wiki/Grep (accessed 3/23/2012).
53. See Appendix 3:P for these Lunyu 7/1 parallels.
54. Citations from the Zhouyi, Lunyu, and Mengzi to Be Found in Pre-Han and Han Texts (2007) is an excellent
resource that lists a number of early Lunyu parallels. However, it is not nearly as comprehensive as the list I was
able to compile using fuzzy searching.
55. Weingarten (2010) also sets aside the figure of Kongzi to focus on Kongzi material itself.
20
symbol throughout the early literature, Kongzi was often quoted without reference to his biography, thereby making “Kongzi”—the man, the symbol, the legend—a projection of the words
attributed to him. Focusing on Kongzi quotations is also useful insofar as the Lunyu compilers
were interested primarily in Kongzi sayings as opposed to Kongzi anecdotes or Kongzi encomia.56 One of the central arguments of this dissertation is that the creation and ascendancy of the
Lunyu Kongzi must be understood against the backdrop of pre-Lunyu Kongzi quotation practice.
The third contribution is more a matter of emphasis than of substance. In contrast to the typical survey of the early period, chapter two’s survey of Kongzi quotation practice begins in 220
CE in the last days of the Eastern Han dynasty and proceeds in reverse chronological order
through the Xin, Western Han, and Qin empires before concluding in the Warring States period.
Such an approach has a few advantages. First, an examination of the Lunyu’s standing in the
Eastern Han, Xin, and late Western Han places the traditional view of the Lunyu in its proper
context. The notion that the Lunyu and the Lunyu alone was compiled by Kongzi’s closest disciples not long after their master’s death was a fitting backstory for a text which, as we will see,
had already been established as the pre-eminent source of Kongzi’s sayings. Second, the contrast
between Kongzi quotation practice in the Eastern Han period on the one hand and and pre-Han
period on the other underscores the Lunyu’s dominance in the former and its profound lack of influence in the latter. Given the tendency to read the Lunyu’s authority back into the early Warring States period, such a contrast is itself a powerful reminder that its authority was not a historical constant. Third, beginning chapter two’s survey in the Eastern Han reverses the tendency to
privilege the seemingly foundational texts of Warring States-era thinkers over sources of early
56. Far and away the most common type of entry in the Lunyu is the standlone zi yue 子曰 (The Master said)
saying. For this point, see p. 84.
21
imperial thought.57 If the Lunyu, historically one of the most cherished sources of Warring States
thought, is actually a product of the Western Han, then there may be a need to reconsider the
chronology of other received texts traditionally dated to the Warring States period in light of the
Han construction of pre-Han texts and ideas. By beginning chapter two’s survey in the Eastern
Han, I hope to encourage other students of early Chinese thought to consider the pre-Han corpus
from a Han perspective.
The fourth respect in which this dissertation differs from previous studies (Kaneto Mamoru’s
and Mark Csikszentmihalyi’s excepted)58 is the extent to which it advances the theory of the Lunyu as a “Western Han text.” Its deficiencies aside, the traditional view of the Lunyu does enjoy
one significant advantage over the revisionist view. Generations of early Chinese scholars (myself included) learned classical Chinese and first studied the masterworks of early China under
the assumption that the Lunyu pre-dates and thus influenced the Warring States masters and later
authors. Not surprisingly, the vast bulk of early China scholarship has largely reinforced the
standard view by interpreting Warring States and Han texts against the Lunyu, thereby giving the
standard view an imposing aura of plausibility. When we encounter a Lunyu parallel in, e.g., the
Mengzi, it simply makes sense to read it as a Lunyu quotation rather than as a Lunyu precursor.
To argue that the Lunyu is a Western Han text, then, one must also demonstrate the plausibility
of reading it as a product of the Western Han context. That is the challenge I take up in chapter
three as well as in a “speculative epilogue” which develops one possible scenario for the Lunyu’s
compilation.
57. See especially Csiksentmihalyi 2006, pp. xx–xxiii, for a discussion of this problem.
58. See especially Kaneto’s (1974) effort to read Lunyu Kongzi sayings with Shiji parallels as later than their Shiji
counterparts.
22
*
*
*
The Lunyu edition used in this dissertation is that of the ICS Ancient Text Concordance Series
(Lunyu zhuzi suoyin 論語逐字索引). Whenever possible, I have relied on a digital edition of the
Sibu congkan 四部叢刊 (Collected Publications of the Four Divisions; hereafter abbreviated
SBCK) for quotations of early received texts. For quotations of Kongzi sayings I have also made
an effort to include citations of the Kongzi jiyu jiaobu 孔子集語校補 (Collected Sayings of
Kongzi, collated and supplemented; hereafter KZJY). Old Chinese reconstructions are those of
Baxter and Sagart 2011. All Shi 詩 (Odes) translations are after Waley 1996. All other translations are my own unless otherwise noted.
23
CHAPTER ONE:
An Introduction to Confucius Sayings
This chapter is an unconventional introduction to Kongzi, not the historical Kongzi but the
Kongzi of Kongzi yue 孔子曰 (Kongzi says), that seemingly ubiquitous presence whose quotations pervade the early corpus. Ironically enough given Kongzi’s familiarity in China and elsewhere, the Kongzi of Kongzi yue remains relatively unknown to the average student of early Chinese thought. This is due in part to the 2,000 year-old preoccupation with the supposedly
authentic Kongzi of the Lunyu, although the logistical difficulties of reading the scattered corpus
of Kongzi sayings has no doubt played a role. Whatever else, the Lunyu Kongzi is an eminently
approachable and manageable Kongzi.
The Kongzi who emerges from this survey is, admittedly, a more protean and frustrating figure. He does not espouse an internally consistent philosophy, nor can his life can be mapped onto
a definite timeline. But grappling with the irreducible messiness of early Kongzi sayings is the
first step towards understanding the nature and evolution of this material in the early period and,
ultimately, breaking out of the Lunyu-centric approach to the study of early Chinese thought.
“Kongzi quotation practice,” my term for the set of informal conventions which dictated how
early authors invoked Kongzi, was far bigger, and far more interesting, than the Lunyu-centric
view would suggest.
At the same time, the sheer abundance of early Kongzi material has led me to make numerous selections of my own in order to highlight what I take to be the most salient features of the
corpus. These selections are by no means unbiased. Throughout this chapter I favor sayings
which are more likely to be early—i.e., those found in pre-Han and early Western Han sources—
24
as well as those which amplify the inadequacies of the traditional view, especially with respect to
the historicity of Kongzi material.59 The traditional answer to the question of why early authors
were interested in quoting Kongzi is simply that Kongzi was a sage of the highest order whose
words conveyed timeless wisdom, and the answer to the question of what early authors (or at
least pro-Kongzi authors) quoted is that they quoted teachings which had been faithfully transmitted by Kongzi’s disciples. As I will argue below, neither of these assumptions is corroborated
by the available evidence. Not only do Kongzi sayings provide very little evidence of faithful
disciple transmission or of an interest in accurately representing the teachings of the historical
Kongzi, a close analysis of the functions of Kongzi sayings reveal Kongzi quotation to have been
a creative practice in which Kongzi sayings interacted dynamically with their contexts. This is in
stark contrast to the Lunyu’s presentation of Kongzi sayings as independent, context-free dicta.
Above all, I hope to show that the value of Kongzi sayings as sources of early Chinese thought
lies not in what they imply about the historical Kongzi, but in the role they played within early
intellectual discourse.
Let us begin with the basics.
What is a Kongzi saying?
The English word “saying” has a more general usage—something that is said or the act of
speaking—and a narrower one—a pithy, self-contained expression. In this dissertation, the term
59. As we will see in chapter two, Kongzi quotation practice in the late Western Han, Xin, and Eastern Han periods
was dominated by the Kongzi sayings of the Lunyu. Consequently, a study of late Western Han and later Kongzi
sayings would tend to reinforce the Lunyu-centric approach to Kongzi.
25
“Kongzi saying” has the latter sense; consequently, not every word attributed to Kongzi will
qualify as a “Kongzi saying.” For instance, when Kongzi is asked at Lunyu 5/8 if his student Zilu
子路 is ren 仁 (humane), his three-word response—“I do not know” (bu zhi ye 不知也)—is not a
“Kongzi saying” because it is not the kind of utterance that is quotable outside of a specific context. “I do not know” only makes sense within a “Kongzi anecdote,” a self-contained story or dialogue within which utterances like “I do not know” acquire meaning.60
Like “saying,” the classical Chinese equivalent yan 言 can refer either to the act of speaking
or to the words that are spoken, but it also has a more specialized usage. At Lunyu 13/15, Kongzi
answers Duke Ding’s 定公 question about “a single yan which can make a state prosper” (一言
而可以興邦): “People have a yan, ‘Being a lord is difficult, and being a subject is not easy’” (人
之言曰：為君難，為臣不易。).61 As Kongzi’s response shows, his interlocutor is not just asking for any string of words. He demands a yan, a memorable expression which can be repeated
and quoted anew. Kongzi yan are a subset of a larger category that I refer to as “Kongzi material,” which includes Kongzi anecdotes as well as “Kongzi testimonia,” or authors’ statements
about Kongzi.
However, early authors’ habitual disrespect for such distinctions cautions against taking the
form of any given piece of Kongzi material as fixed in the minds of early authors. In particular,
examples of saying→anecdote variability abound in early texts, as when a conversation between
Kongzi and his disciple Zigong 子貢 in the Lüshi chunqiu 呂氏春秋 (Mr. Lü’s Annals) appears
60. Not surprisingly, no early author seems to have quoted Kongzi with the words “I do not know.” A passage from
Xunzi book 29 (“Zi dao 子道” [“The Way of the Son”]) has Kongzi admitting his ignorance (wu bu zhi ye 吾不知
也) in response to a student’s question; however, it rationalizes Kongzi’s response by pointing out an error in the
question. See Xunzi 29 (SBCK 29/11b-12a; Kongzi jiyu jiaobu [hereafter KZJY] 9.31/p. 184).
61. SBCK 13/5b.
26
at Lunyu 7/2 as an independent saying,62 or when Kongzi’s response to a question from Ziyou 子
游 at Lunyu 1/15 appears in the Yantie lun 鹽鐵論 (Salt and Iron Discourses) and the “Fang ji 坊
記” (“Embankment Record”) chapter of the Liji 禮記 as independent sayings.63 Lifting a Kongzi
saying from its context was apparently widespread enough to have been criticized by the author
of the “Tan gong 檀弓” (“Sandalwood Bow”) chapter of the Liji 禮記 (Record of Ritual). There
Youzi 有子, a figure whose sayings also appear in the Lunyu, complains about the tendency of
others to quote what the Master said without also mentioning “what the Master said it about”
(夫子有為言 fuzi you wei yan), because the Master’s words by themselves do not necessarily
communicate his full intention.64 Such decontextualization parallels the practice of duan zhang
qu yi 斷章取義 (“breaking off stanzas and taking their meaning”) in Shi 詩 quotation practice,
whereby quoters decoupled rhymed couplets from their original contexts in order to imbue them
with new meanings and apply them to new situations.65 In both Kongzi and Shi quotation practice, an author’s immediate rhetorical demands often trumped any concern to faithfully reproduce the original context of the words being quoted.
Early authors also seem to have invented contexts for Kongzi sayings and turned them into
Kongzi anecdotes, as when the probably much later Kongzi jiayu 孔子家語 (Sayings of the
House of Kongzi) turned two Zhongni yue sayings from the earlier Zuozhuan into conversations
between Kongzi and Zigong 子貢.66 Youzi’s complaint from the “Tan gong” provides one ratio-
62. See Lüshi chunqiu “Zun shi 尊師” (KZJY 1.23/p. 8; SBCK 4/7b) and Lunyu 7/2 (SBCK 7/1a).
63. See Lunyu 1/15 (SBCK 1/6a) and Liji 30 (SBCK 15/12b; KZJY p. 546).
64. Liji 3 (SBCK 2/15a; KZJY pp. 521–2). See p. 46 below for a translation of of this episode.
65. The locus classicus of this term is Zuozhuan Xiang 28 (SBCK 8/17a).
66. See Zuozhuan Xiang 25 (SBCK 17/14b; KZJY p. 415) and Kongzi jiayu (SBCK 9/18a), and Zuozhuan Xi 28
(SBCK 7/8b; KZJY p. 413) and Kongzi jiayu (SBCK 10/1a). The former saying is discussed on p. 55 below.
27
nale for such inventions: without its proper context, a Kongzi saying might be used in such a way
as to undermine Kongzi’s reputation as a supremely wise and virtuous figure. Another motivation is supplied by a saying preserved in the Shiji postface: “The Master said, ‘My desire to
record empty words does not compare to my desire to see a person’s deeds in all their immediacy
and vividness’” (子曰：“我欲載之空言，不如見之行事之深切著明也。).67 Sima Qian
echoes this sentiment in a comment on his biography of Kongzi: “When I read the writings of
Master Kong, I imagine that I see him as a man” (余讀孔氏書，想見其為人。).68 Apparently, it
was this interest in Kongzi “as a man” (qi wei ren 其為人) which led Sima Qian to situate
Kongzi’s potentially “empty words” within a biographical framework which is the most “vivid”
portrayal of Kongzi known from the early period. In the Shiji, it is Kongzi’s deeds that give the
true measure of the man.
Another kind of variability is evident in a comparison between Mengzi 7B/37 and Lunyu
13/21, in which Mengzi’s description of Kongzi in the former is represented as a zi yue saying in
the latter:
<Mengzi 7B/37>Mengzi said, “Kongzi could not associate with [men of] the middle way...”
孟子曰：孔子不得中道而與之...
<Lunyu 13/21>The Master said, “When one cannot associate with [men of] the middle way...”
子曰：不得中行而與之...
67. Shiji 130.3297 (KZJY 5.65, p. 86). The context of this attribution in the Shiji (as well as in the Chunqiu fanlu
book 17, “Yu xu 俞序” [SBCK 6/3b; KZJY 5/67; p. 87]) is Kongzi’s supposed authorship of the Chunqiu 春秋
(Spring and Autumn Annals). The same idea is echoed in a Kongzi saying at Shiji 47.1947 and again in Huainanzi
book 13 (SBCK 13/3b): “Reciting the Shi and Shu of the former kings is not as good as hearing their words. Hearing
their words is not as good as grasping why they spoke. Grasping why they spoke is to say what cannot be said” (誦
先王之詩、書，不若聞得其言，聞得其言，不若得其所以言，得其所以言者，言弗能言也。).
68. An anecdote preserved in the “Kongzi shijia” (Shiji 47.1925) projects this desire onto Kongzi himself as he
learns to play the zither. After his teacher tells him that he has mastered a certain piece of music and can go on to the
next, Kongzi insists on practicing the old tune until he has “grasped [its composer] as a man” (得其為人). The
composer turns out to have been King Wen 文王.
28
Modern editors tend to insert quotation marks around the Lunyu parallel in the Mengzi on the assumption that Mengzi knew, and thus had to have been quoting, the Lunyu.69 But the fact remains
that the Mengzi passage is not a direct quotation, and that the Mengzi’s description of Kongzi
reads very differently from the prescription that the Lunyu attributes to him. Whatever the precise relationship between Mengzi 7B/37 and Lunyu 13/21 (see p. 199), the boundary between
Kongzi testimonia and Kongzi sayings may have been equally porous.
These observations about the fluidity of different types of Kongzi material expose the arbitrariness of my own terminology. Consider again the example of Lunyu 13/15. From the perspective of Kongzi and Duke Ding, the only yan of interest is the anonymous saying “Being a lord is
difficult, and being a subject is not easy.” For the Lunyu’s audience, however, Kongzi’s responses to Duke Ding might have also been read as “sayings” in their own right. On the other hand,
the existence of parallel dialogues in other early texts would suggest that Lunyu 13/15 circulated
not simply as a Kongzi saying but also as a Kongzi dialogue or anecdote.70 Cases like Lunyu
13/15 simplify matters somewhat by suggesting their own terminology—yan 言 (saying)—and
drawing their own boundaries. More often than not, however, parsing complex Kongzi anecdotes
and labeling the “Kongzi sayings” therein is simply my way of marking what I take to be the
most quotable elements of a given passage.
My decision to focus on the history of Kongzi sayings and not Kongzi anecdotes or Kongzi
material generally reflects my interest in how early authors used Kongzi’s words and, by extension, how we ourselves might use Kongzi’s words as sources of early Chinese thought. But there
69. See, e.g., Mengzi zhengyi 29.1026. D.C. Lau (1970, p. 202) also translates this line as a Kongzi quotation:
“Mencius answered, ‘Confucius [said], “If one fails...”’”
70. I have found no such parallels for Lunyu 13/15.
29
are also significant differences with respect to the circulation of Kongzi sayings versus Kongzi
anecdotes and dialogues in the early period. Consider the story of Kongzi’s travails “between
Chen and Cai” (Chen Cai zhi jian 陳蔡之間), perhaps the most widely attested Kongzi anecdote
in the early literature. According to the Shiji version of the story, the leaders of Chen and Cai
sent soldiers to surround Kongzi and his disciples in order to prevent Kongzi from reaching Chu
楚, becoming an adviser to the Chu king, and thereby abetting Chu’s hegemony.71 As seen in Appendix 1:A, a dozen variations of the story can be found in at least eight texts through the Western Han period.72
One might be tempted to take these dozen parallels as evidence that the historical Kongzi really did encounter difficulties between Chen and Cai. However, for all their agreement on the
broad outlines of the story, early authors do not seem to have agreed on the details because the
Kongzi sayings which are the centerpieces of these episodes look substantially different from
version to version.73 Even among the versions at Lüshi chunqiu 14/6, Zhuangzi book 28, Hanshi
waizhuan book 7, and Shuiyuan book 17, all of which feature Kongzi’s disciples questioning the
righteousness of their mission and Kongzi scolding them for their doubts, the core Kongzi sayings differ in both length and content. This would suggest that “between Chen and Cai” stories
represent a sub-genre of Kongzi anecdote, in Jeffrey Riegel’s words a “historical romance,”
whose details early authors were free to vary as they saw fit.74 Thus the Mozi author who was
71. Shiji 47.1930.
72. For a collection of passages related to Kongzi’s misadventures between Chen and Cai, see Shanghai bowu guan
cang Chu zhushu vol. 8, pp. 130–136. See also Makeham 1998 for study of various extant versions of the legend.
73. See also Haupt 2006, p. 18, and Makeham 1998.
74. Riegel 1986, p. 13. In the same article, Riegel argues that these “literary remains of Confucius’s life consist of
bits and pieces of ancient poetry which in their origins had nothing to do with Confucius and even predated him” (p.
14). See also Haupt 2006, p. 88, for the observation that the story of Kongzi’s wanderings roughly parallels that of
Chong’er 重耳, the future Duke Wen of Jin 晉文公, in the Zuozhuan.
30
hostile to Kongzi could turn it into a story about Kongzi’s hypocrisy, and the Zhuangzi book 20
author could turn it into an encounter scene with a mysterious sage figure who instructs Kongzi
in “the Way of not dying” (bu si zhi dao 不死之道). If the “between Chen and Cai” legend provided a narrative framework flexible enough to accommodate whatever Kongzi sayings an author required, then it is the sayings themselves which reveal those authors’ attitudes towards
Kongzi and the boundaries of Kongzi quotation practice.
What I refer to in this dissertation as “the corpus of Kongzi sayings” is, admittedly, an artificial construct. I suspect that no early author conceived of a complete collection of extant Kongzi
material as, for instance, Sun Xingyan 孫星衍 did in the early 19th century when he compiled
the Kongzi jiyu 孔子集語 (Collected Sayings of Kongzi), a compendium of Kongzi material outside of the Lunyu and a few other canonical sources.75 To the contrary, judging from the titles of
early collections of Kongzi material, early compilers tended to draw lines between authoritative
and less authoritative sayings. The Lunyu contained “selected” or “assessed” (lun 論) sayings,
the Kongzi jiayu “sayings from the house of Kongzi,” and the Kong congzi 孔叢子 “Kong family
masters” sayings.76 At the risk of anachronism, my focus on a complete corpus of Kongzi sayings
is methodologically useful insofar as it frees me from having to make problematic distinctions
between authentic and inauthentic, or more interesting and less interesting, sayings.
75. In a preface to the work, fellow compiler Yan Kejun 嚴可均 (1762–1843) lists several texts whose Kongzi
sayings Sun Xingyan excluded from the collection on the grounds that they were already widely available. These
texts were the Zhouyi, Liji, Chunqiu Zuozhuan, Xiaojing, Lunyu, Mengzi, Kongzi jiayu, Kong congzi, and the
“Kongzi shijia” and “Zhongni dizi liezhuan.” See Kongzi jiyu jiaobu 1998, pp. 1–2.
76. It is interesting that, of these three collections, the Lunyu is the only text whose title does not invoke genealogy
to assert its authenticity. Although not an early compilation, Liang Emperor Wu’s 梁武帝 (r. 502–549) no longer
extant Kongzi zhengyan 孔子正言 (“Correct Sayings of Kongzi”) continued this practice of privileging some
sayings over others. The earliest reference to this work is Suishu 32.937.
31
How are Kongzi sayings identified?
Kongzi sayings are typically marked by the phrase Kongzi yue 孔子曰 (“Kongzi says/said”) or
some variation thereof. The most common variations substitute a different name for Kongzi (e.g.,
Zhongni 仲尼 or Kong Qiu 孔丘)77 and/or a synonym for yue 曰 (e.g., yun 云 or cheng 稱). The
phrase Lunyu yue 論語曰 (“The Lunyu says”) introduces Kongzi sayings found in the Lunyu,
which is one of the very few sources of Kongzi sayings privileged in this way.78 Early authors
occasionally introduced sayings with the rhetorical question Kongzi bu yun hu 孔子不云乎
(“Did not Kongzi say...?”), a formulation which seems to assume the audience’s familiarity with
the saying in question.79 Another common variation is the phrase Kongzi wen zhi yue 孔子聞之
曰 (“Kongzi heard [or learned of] this and said”), which introduces Kongzi sayings as comments
on historical (or pseudo-historical) anecdotes. Occasionally, early authors seem not to have cared
whether they were quoting Kongzi or some other source so long as the quotations were backed
by tradition, as shown by their use of the phrase “zhuan yue” 傳曰 (“a tradition has it that”) to introduce sayings that appear elsewhere as Kongzi sayings.80 Authors also occasionally quoted
what Kongzi “wrote” (shu 書) rather than what he said, especially in the context of Kongzi’s
purported authorship of the Chunqiu chronicle, as in this example from the Hou Hanshu 後漢書
77. The earliest text to explicitly tie these names to a single figure is the “Kongzi shijia” (Shiji 47.1905–1948),
which explains that “Kong” was his family name (xing 姓), “Qiuyun 丘云” his given name (ming 名), and “Zhongni
仲尼” his cognomen (zi 字). Earlier texts like the Lüshi chunqiu and Zuozhuan corroborate the “Kongzi shijia” by
using “Kongzi,” “Zhongni,” and “Qiu” interchangeably in the same passages. Christiane Haupt (2006, p. 50) has
noted that all of the Zhongni yue sayings in the Zuozhuan appear in the Kongzi jiayu as Kongzi yue.
78. For other examples, see my discussion in chapter two (p. 152) of Kongzi quotation in the Baihu tongyi 白虎通
義 (The General Meaning of the White Tiger Hall Discussions).
79. See, e.g., Emperor Ai’s 哀帝 edict from 7 BCE discussed on p. 100 below.
80. See, e.g., Emperor Xuan’s 宣帝 edict from 66 BCE at Hanshu 8.250. According to Schaberg (2001, p. 79),
“[T]he speakers and writers who built inherited language into speeches during the Eastern Zhou were often
uncertain of or indifferent to the provenance of that language.” See also Schaberg 2005, pp. 4–6.
32
(History of the Later Han): “When Kongzi authored the Chunqiu he wrote ‘the first month’ to
show his respect for the start of the year” (孔子作春秋，書正月者，敬歲之始也。).81 But
these examples do not begin to exhaust the variations one finds in Kongzi quotation markers.82
A substantial subset of Kongzi sayings are prefaced by the more ambiguous zi yue 子曰 and
fuzi yue 夫子曰 markers. In texts like the Lunyu, Kongzi jiayu 孔子家語, and Kong congzi 孔叢
子 whose interest in Kongzi is obvious, zi and fuzi refer unambiguously to “The Master,” i.e.,
Kongzi. But the zi is not always synonymous with Kongzi. In the Mengzi 孟子, Mozi 墨子, and
other master texts, zi and fuzi refer unambiguously to other masters.83 Even in the Lunyu, Kongzi
once refers to an interlocutor’s lord as fuzi.84
In many other instances the referent of zi yue/fuzi yue is less obvious.85 For example, neither
of the manuscript versions of the “Ziyi 緇衣” (“Black Robes”) nor the version included in the received Liji 禮記 (Ritual Records) mentions the name of the master whose sayings pervade that
text.86 Similarly, not one of the thirty zi yue sayings in the received Zhouyi 周易 (Zhou Changes)
81. Hou Hanshu 30.1071. Other examples include Shiji 43.1791 (“Kongzi...wrote in the Spring and Autumn
Annals...” 孔子...書春秋曰), Hanshu 93.3741 (“Zhongni wrote [in Lunyu 16/4], ‘He who loses befriends three kinds
of people’” 仲尼著損者三友), Hanshu 99.4089 (“Kongzi wrote in the Classic of Filial Piety” 孔子著孝經曰), and
Hou Hanshu 61.2029 (“Kongzi criticized him and wrote” 孔子譏之書曰). However, my impression is that such
examples are relatively rare.
82. In 20 CE Wang Mang 王莽 quoted Kongzi as “His Excellency the Resplendent Ni” (Xuan Ni gong 宣尼公), a
shorthand for the posthumous title granted to Kongzi in 1 CE, “His Excellency the Resplendent Ni of Baocheng
[i.e., Recompense for Perfection]” (Baocheng xuan Ni gong 褒成宣尼公). For Wang Mang’s edict see Hanshu
99.4160 and p. 160 below; for the origin of this title, see Hanshu 12.351.
83. See, e.g., Mengzi 2B/13 (SBCK 4/15b), although the fuzi yue saying in that passage (“A noble man neither
resents Heaven nor finds fault with other men” [君子不怨天，不尤人。]) also appears at Lunyu 14/35.
84. Lunyu 14/25.
85. Schaberg 2005, p. 15: “[T]he attribution of sayings to a zi is widespread, even in texts not explicitly associated
with Confucius, and one wonders what other translations of zi yue 子曰 are possible, even correct: ‘a master said’;
‘my teacher said’; perhaps ‘a teacher would say.’ Depending on how one translates the framing gesture for
aphorisms in the Lunyu, historical questions concerning retrospective attribution of newly fabricated sayings may be
less pressing.”
86. The authorship of “Ziyi” was apparently a controversial topic even in the early medieval period, with Liu Xian
劉獻 (434–489 CE) attributing it to Kongzi’s disciple Gongsun Nizi 公孫尼子 and Shen Yue 沈約 (441–513) to
33
includes any information tying that zi to Kongzi or any other figure.87 However, as with many
early texts of anonymous origin, an author was eventually found for them, as when the authors of
the Hanshu “Wuxing zhi 五行志” (“Treatise on the Five Phases”) and Qianfu lun 潛夫論 (Discourse of a Hidden Master) in the Eastern Han period unambiguously attributed a zi yue saying
from the Xici zhuan 繫辭傳 (Commentary on the Appended Phrases) to Kongzi.88
Complicating matters further, comparisons among parallel versions of Kongzi sayings in early texts indicate that the distinction between Kongzi yue and zi yue sayings was not consistently
maintained. Two striking examples of this phenomenon are the Dingzhou Lunyu manuscript and
the “Kongzi shijia 孔子世家” (“Hereditary House of Kongzi”) chapter of the Shiji 史記. Not
only do a handful of passages from the Dingzhou Lunyu manuscript have Kongzi yue where the
received version has zi yue and vice versa,89 in every instance where the “Kongzi shijia” overlaps
with the Lunyu the former has Kongzi yue and the latter has zi yue.90 Clearly, early authors did
not adhere to a single standard when quoting either “Kongzi” or “The Master” (or perhaps “a
master”). Given the frequency of Kongzi yue/zi yue overlap, in this dissertation I follow Sun
Xingyan’s lead and include zi yue sayings alongside Kongzi yue sayings, except where the con-
Zisizi 子思子. Shen Yue 沈約 believed that the Liji was reconstituted following the Qin bibliocaust by Han Ru 儒
who pieced the text together from disparate sources, “none of which were the words of the sage” (皆非聖人之言).
For Shen Yue’s argument, see Suishu 13.288. For Lu Deming’s 陸德明 (556–627) apparent endorsement of Liu
Xian’s view, see Liji zhengyi (SBCK 17/9b).
87. The one possible exception is a reference to Yan shi zhi zi 顏氏之子 (“the son of the Yan clan”), i.e., Kongzi’s
favorite disciple Yan Hui 顏回, in the “Xici zhuan” (SBCK 8/5b). However, the passage in question seems to treat
Yan Hui as a famous exemplar rather than as a personal acquaintance of Kongzi. On the other hand, it is clear from
some of the Mawangdui Yi materials that this master was felt to be Kongzi.
88. For the original saying from the Xici zhuan, see SBCK 7/5b; KZJY p. 445. For the Hanshu and Qianfu lun
versions, see Hanshu 27.1376 and Qianfu lun 32 (SBCK 8/9a–b).
89. See Weingarten 2010 (pp. 45–48) for a more in-depth discussion of this issue, as well as the effort by some
scholars to use the zi yue/Kongzi yue distinction to date the layers of the Lunyu.
90. This likely has something to do with the fact that there are many “masters” (zi 子) in the Shiji, so that referring
to Kongzi simply as “The Master” would have invited confusion.
34
text demands otherwise. By this I do not mean to imply that all zi yue sayings were originally
meant to be read as Kongzi sayings. Methodologically, one simply cannot appreciate the distribution and variability of Kongzi sayings if one excludes the overlapping corpus of zi yue sayings.
Statements attributed to the anonymous junzi 君子 (noble man) have also been read as
Kongzi sayings by commentators from the Han period onwards. For instance, Wang Chong 王充
(27–100 CE) made this connection explicit when he observed of a junzi saying in the Gongyang
that “‘junzi’ refers to Kongzi” (君子者，孔子也),91 a conclusion which the Gongyang itself
seems to encourage.92 Whether pre-Han authors would have agreed with the association of
Kongzi with the junzi is an open question. In the absence of evidence to the contrary, it is perhaps best to understand the junzi figure not as a specific persona but as a generic ideal type, or
perhaps as a reference to a specific individual whose identity was not recorded.
Another problematic feature of Kongzi quotation practice (and early quotation practice generally) is the tendency of early authors to use Kongzi sayings without marking them as such. In
the following passage from Wang Chong’s 王充 (27–100 CE) Lunheng 論衡, the implicit quotation of Lunyu 3/5 is obvious given both the title of the treatise—“Wen Kong” 問孔 (“Interrogating Kongzi”)—and the use of the word yan 言 (“[this] says”) to introduce a clarifying comment
on the quoted text:
I ask: What gave rise to Kongzi’s desire to go among the barbarians? It was because the Way
was not followed in the central states that he wished to go among the barbarians. But if the
Way was not followed in the central states, how could it be followed in barbarian lands?
“Barbarians with a ruler cannot compare to Xia peoples without one” says that what is
difficult for barbarians is easy for the Xia peoples. If the Way cannot be followed by those for
whom it is easy, how can it be followed by those for whom it is difficult?
問之曰：孔子欲之九夷者，何起乎？起道不行於中國，故欲之九夷。夫中國且不行，安
91. Lunheng 27 (SBCK 8/15b).
92. The final entry of the Gongyang (SBCK 12/9b) follows a number of Kongzi sayings with the question, “Why
did the junzi make the Chunqiu?” (君子曷為為春秋).
35
能行於夷狄？夷狄之有君，不若諸夏之亡。言夷狄之難，諸夏之易也。不能行於易，能
行於難乎？93 Aside from the obvious word-for-word Lunyu parallel, the decision to insert quotation marks
around the line “Barbarians with a ruler...” is also justified by Wang Chong’s frequent use of the
Lunyu yue 論語曰 (The Lunyu says) quotation marker to preface word-for-word Lunyu quotations elsewhere in the Lunheng. There is little reason to doubt that Wang Chong knew and quoted a text closely resembling the received Lunyu.
More often than not, however, the identification of implicit quotations is fraught with uncertainty because it depends on one’s assumptions about the texts a given author might have been
familiar with.94 Consider the following line attributed to Sima Jizhu 司馬季主, a Chu 楚 diviner
profiled in the Shiji “Rizhe liezhuan 日者列傳” (“Biography of the Hemerologists”): “To transmit without originating is the principle of the noble man” (述而不作，君子義也。).95 For anyone familiar with the Lunyu, the first four characters immediately recall Lunyu 7/1: “Transmitting without initiating, trusting in and loving the ancient, I humbly compare myself to Old Peng”
(述而不作，信而好古，竊比於我老彭). Nevertheless, these four characters are probably not a
quotation of Lunyu 7/1. Even if we assume that Sima Qian had access to a Lunyu text, the characters following shu er bu zuo in this particular passage bear no resemblance to the Lunyu version.96 Moreover, nowhere else in the Shiji (or in any other Western Han text, for that matter) is
93. Lunheng 28 (SBCK 9/10b).
94. Schultz 1999, p. 142: “[Q]uotation is an extremely complex phenomenon. Sometimes it is clearly marked by an
introductory formula but more often it is left unmarked and its presence can be detected only by contextual features,
such as archaic vocabulary, grammatical awkwardness, abrupt style change, or simply the recognition that the same
passage appears elsewhere. The assessment is as difficult as the detection, for the quoter can be embracing,
questioning, modifying, or even rejecting the original meaning of the passage.”
95. Shiji 127.3218.
96. See pp. 182–187 for a discussion of the relationship between the Lunyu and the Shiji.
36
shu er bu zuo 述而不作 explicitly attributed to Kongzi.97 As a general rule, one is justified in
treating a passage as an implicit quotation if there is reason to think that (a) a given author was
familiar with the source text in question, and (b) the candidate quotation closely parallels the version found in that particular source text.
Determining the precise length of a Kongzi saying can be equally problematic because early
authors did not adhere to a single standard for punctuating their quotations. Although the length
of most sayings is clear from the context, early texts abound with more frustrating examples.
One of these is the aforementioned “Zi yi,” which consists of a series of zi yue sayings capped
with quotations from canonical texts, as in the first section of the received version:
The Master said: If one loves accomplishment as in [the Shi] “Black Robes,” and if one detests
the detestable as in [the Shi] “Chief Eunuch,” then ranks are not belittled and the people are
spurred to be admiring. Although punishments are not used the people all submit. One of the
Greater Elegentiae says, “Modeling themselves on King Wen, the myriad states act
faithfully.”
子曰：好賢如緇衣，惡惡如巷伯，則爵不瀆而民作愿，刑不試而民咸服。大雅曰：儀刑
文王，萬國作孚。98
Does the zi yue quotation end before or after the Shi quotation marker? How we answer that
question has important consequences for our understanding of the text. If before, then the “Zi yi”
author deserves the credit for pairing it with the zi yue saying. One might then infer that the “Zi
yi” author set out to buttress some master’s teachings by pointing out certain convergences with
the Shi. On the other hand, if the zi yue attribution includes the Shi quotation, then it is the unnamed master who demonstrates his mastery of the Shi.99 When dealing with Kongzi yue sayings
97. For Lunyu 7.1 parallels, see Appendix 3:P.
98. Liji zhengyi (SBCK 17/10a; KZJY p. 560).
99. My own view, which is based on the observation that certain recurring patterns fall between the zi yue quotation
markers and the capping Shi and Shu quotations, is that the “Zi yi” author inserted a layer of comments between the
zi yue material and the capping quotations. For this argument, see Hunter & Kern, forthcoming.
37
and attributed language generally, how one punctuates such passages has important interpretive
consequences.
Where are Kongzi sayings found?
The following numbers provide only a rough sense of the scale of the corpus of Kongzi sayings.
The Kongzi jiyu jiaobu 孔子集語校補 (Collected Sayings of Kongzi, collated and supplemented), a substantially revised edition of Sun Xingyan’s original collection, contains over 1,700 entries culled from received and discovered texts from the early period.100 Adding this to this the
500 Lunyu entries, 360 Kongzi jiayu entries, and 140 Kong Congzi entries yields a total of roughly 2700 Kongzi-related entries from the most important collections of early Kongzi material.
This figure does not even include all of the Kongzi yue sayings found in manuscripts to date,
quotations of Lunyu Kongzi sayings, or instances like the Lunheng’s implicit quotation of Lunyu
3/5 discussed above. All told, early texts preserve somewhere between 3000 and 4000 instances
in which an author explicitly attributed a piece of text to Kongzi.
Texts that cite Kongzi can be grouped into the following eight categories (with sources of zi
yue sayings marked with an asterisk*):
● Collections of Kongzi material. Compilations of Kongzi material are far and away the largest
sources of Kongzi sayings. The most important of these are the Lunyu, Kongzi jiayu 孔子家語
(Sayings of the House of Kongzi), and Kong Congzi 孔叢子 (Kong Family Masters Anthology);
the “Kongzi shijia 孔子世家” (“Hereditary House of Kongzi”) and “Zhongni dizi liezhuan 仲尼
100.“Entry” is my (admittedly imprecise) term for a self-contained unit of Kongzi material and could refer to a
saying, an anecdote or dialogue, with multiple Kongzi attributions, or a statement about Kongzi.
38
弟子列傳” (“Biographies of Zhongni’s Disciples) chapters of the Shiji 史記 (Grand Scribe’s
Records); and the Liji 禮記 (Records of Ritual) chapters “Zengzi wen 曾子問” (“Zengzi asked”),
“Ai Gong wen 哀公問” (“Duke Ai asked”), “Zhongni yanju 仲尼燕居” (“Kongzi at leisure”),
and “Kongzi xianju 孔子閒居” (“Kongzi at rest”). Four additional Liji chapters—“Fangji 坊記”
(“Embankment Record”)*, “Zhongyong 中庸” (“Doctrine of the Mean”)*, “Biaoji 表記” (“Exemplary Record”)*, and “Ziyi 緇衣” (“Black Robes”)* chapters—are major sources of zi yue
sayings.
● The Classics. As one would expect from the traditional account of Kongzi’s life, Kongzi sayings do not appear in texts traditionally dated to the Western Zhou and early Spring and Autumn
periods, i.e., the Shi 詩 (Odes), Shu 書 (Documents), and the core layers of the Zhouyi 周易
(Zhou Changes). Although Kong Qiu is mentioned only once in the text of the Chunqiu 春秋
(Spring and Autumn Annals),101 Kongzi is quoted extensively in the Zuozhuan 左傳 (Zuo Traditions) and to a much lesser extent in the Gongyang 公羊 and Guliang 榖梁 traditions. Of the san
li 三禮 (three ritual classics) texts, Kongzi sayings appear most frequently in the Liji (see above).
The Yili 儀禮 (Etiquette and Rituals) includes just one Kongzi yue saying. The Xiaojing 孝經
(Classic of Filial Piety), traditionally attributed to Kongzi’s disciple Zengzi 曾子, consists of a
lengthy dialogue between Zengzi and Kongzi. The following Da Dai liji 大戴禮記 (The Greater
Dai’s Ritual Records] chapters also contain Kongzi material: “Zhu yan 主言” (“Mastering
words”), “Ai gong wen wu yi 哀公問五義” (“Duke Ai asked about the five principles”), “Ai
gong wen yu Kongzi 哀公問於孔子” (“Duke Ai asked Kongzi”), “Li cha 禮察” (“Ritual investigated”), “Wei jiangjun Wenzi 衛將軍文子” (“Wenzi, General of Wei”), “Wu di de 五帝德”
101.This is the famous line at Ai 16 (SBCK 30/7a) that announces Kongzi’s death: “Kong Qiu passed away” (Kong
Qiu zu 孔丘卒).
39
(“The virtues of the five emperors”), “Quan xue 勸學” (“Encouraging learning”), “Zizhang wen
ru guan 子張問入官” (“Zizhang asked about entering office”), “Qian sheng 千乘” (“One-thousand chariots”), “Si dai 四代” (“The four ages”), “Yu dai de 虞戴德” (“Shun donned his
virtue”), “Gao zhi 誥志” (“Proclaiming intent”), “Xiao bian 小辨” (“Lesser discriminations”),
“Yong bing 用兵” (“Using the military”), “Shao xian 少閒” (“A little rest”), and “Yi ben 易本”
(“The root of the Changes”).
● Commentaries. Kongzi and zi yue sayings played a prominent role in early commentaries:
◦ Yi 易 (Changes) texts. These include the “Wen yan 文言” (Patterned Words)* and “Xici
zhuan 繫辭傳” (Commentary to the Appended Phrases)* of the received Zhouyi as well as
the “Er san zi wen 二三子問” (“The Disciples Asked”), “Yi zhi yi 易之義” (“The Meaning
of the Changes)*, “Yao 要” (“Essentials”)*, “Miaohe 繆和”*, and “Zhao li 昭力”* texts
from the Mawangdui 馬王堆 find.
◦ Shi 詩 (Odes) texts: the Hanshi waizhuan 韓詩外傳 (Outer Commentary on the Han Odes),
the Shanghai Museum “Kongzi shilun 孔子詩論” (“Kongzi’s Discussion of the Odes”), and
both the Mao 毛 and Zheng Xuan 鄭玄 (127–200) commentaries on the Shi.
◦ Shu 書 (Documents) texts: the Shangshu dazhuan 尚書大專 (Great Commentary to the Exalted Documents) and the various Shu comments preserved in the “Lun shu 論書” (“Discussing the Shu”) chapter of the Kong Congzi.
◦ Chunqiu 春秋 (Spring and Autumn Annals) texts: all three Chunqiu 春秋 commentaries (the
Zuo zhuan 左轉, Gongyang zhuan 公羊傳, and Guliang zhuan 榖梁傳).
◦ Li 禮 (Ritual) texts: Zheng Xuan’s commentaries on the Zhouli 周禮 (Zhou Rituals), Yili, and
Liji.
◦ Other extant commentaries from the Eastern Han period. These include Wang Yi’s 王逸 (fl.
40
c. 115 CE) Chuci 楚辭 (Verses of Chu) commentary, Zhao Qi’s (108–201 CE) Mengzi 孟子
commentary; He Xiu’s 何休 (129–182) Gongyang commentary; the Xu Shen 許慎 (55–149
CE)/Gao You 高誘 (168–212) commentaries on the Huainanzi 淮南子; and Gao You’s Lüshi
chunqiu 呂氏春秋 (Mr. Lü’s Annals) commentary. The Baihu tongyi 白虎通義 (General
Meaning of the White Tiger Discussions), the summary of a conference convened by Emperor Zhang 章帝 (57–88 CE) in 79 CE in order to discuss “agreements and disagreements
among the Five Classics” (wujing tongyi 五經同異), also contains numerous Kongzi sayings.
◦ “Prognostication” (chen 讖) commentaries. “Yi chen 遺讖” (“Remnant apocrypha”), the
thirteenth chapter of the Kongzi jiyu, contains a number of fragments from apocryphal commentaries to the Yi, Shu, Shi, Chunqiu, Xiaojing, and Lunyu.
● Histories (shi 史). Sima Qian’s (d. c. 87 BCE) Shiji 史記 (Grand Scribe’s Records), Ban Gu’s
(32–92) Hanshu 漢書 (Han Documents), and Fan Ye’s (398–445) Hou Hanshu 後漢書 (Later
Han Documents) preserve a large number of sayings in memorials (zou 奏), edicts (zhao 詔/zhi
制), monographs (shu 書), treatises (zhi 志), biographical narratives (liezhuan 列傳), and in the
historians’ own comments. The most important sources of Kongzi sayings in the early histories
are the Shiji chapters “Kongzi shijia 孔子世家” (“Hereditary House of Kongzi”) and “Zhongni
dizi liezhuan 仲尼弟子列傳” (“Biographies of Zhongni’s Disciples”).
● Eclectic (za 雜) compilations. Large compilations like the Lüshi chunqiu 呂氏春秋 (Mr. Lü’s
Annals), Huainanzi 淮南子, Shuiyuan 說苑 (Garden of Persuasions), and Xinxu 新序 (New
Arrangement) preserve a wealth of Kongzi material in various formats.
● The texts of the masters (zi 子). Not surprisingly, Kongzi sayings are most prevalent in master
texts like the Mengzi 孟子and Xunzi 荀子 that are associated with the Ru 儒 tradition. Of these
texts, the final five chapters of the Xunzi (“You zuo 宥坐” [“The Warning Vessel on the Right”],
41
“Zi dao 子道” [“The Way of Sons”], “Fa xing 法行” [“A Model for Conduct”], “Ai gong 哀公”
[“Duke Ai”], “Yao wen 堯問” [“Yao Asked”]) contain a disproportionate number of Kongzi sayings (roughly fifty) in a list format that somewhat resembles the Lunyu. Other Ru texts which
quote Kongzi include the Yanzi chunqiu 晏子春秋 (Master Yan’s Annals), Lu Jia’s 陸賈 Xinshu
新書 (New Text), Jia Yi’s Xinyu 新語 (New Sayings), Dong Zhongshu’s 董仲舒 Chunqiu fanlu
春秋繁露 (Luxuriant Dew of the Spring and Autumn Annals), Huan Tan’s 桓譚 Xinlun 新論
(New Discourses), Ying Shao’s 應劭 Fengsu tongyi 風俗通義 (Comprehensive Assessment of
Contemporary Customs), and Xu Gan’s 徐幹 Zhonglun 中論 (Discourses that Hit the Mark).
One also finds Kongzi sayings in a number of non-Ru texts, including the Mozi 墨子, Han Feizi
韓非子, and Zhuangzi 莊子. Kongzi sayings are noticeably absent from several other zi texts,
e.g., the Laozi 老子, Shangjun shu 商君書 (Writings of Lord Shang), and Guanzi 管子. Other
notable sources of Kongzi material are Yang Xiong’s 揚雄 (53 BCE–18 CE) Fayan 法言 (Model
Sayings), a text modeled after the Lunyu, and Wang Chong’s 王充 (27–100) treatise “Wen Kong
問孔” (“Questioning Kongzi”), which critiques roughly a dozen passages from the Lunyu.
● Manuscripts. Whereas the vast majority of received Kongzi sayings are preserved in compilations of various kinds, Kongzi-related manuscripts tend to be shorter, unincorporated texts. The
most significant finds are the Dingzhou 定州 Lunyu, an early manuscript version of the Lunyu
dated to 55 BCE whose fragments correspond to roughly half of the entries in the received Lunyu; the Dingzhou “Rujia zhi yan 儒家之言” (“Sayings of the Ru”) and “Ai gong wen wu yi 哀公
問五義” (“Duke Ai asked about the five principles”) texts;102 the Shanghai Museum “Kongzi
shilun 孔子詩論” (“Kongzi’s Discussion of the Odes”), “Min zhi fumu 民之父母” (“Father and
102.For “Rujia zhe yan,” see Hebei-sheng wenwu yanjiusuo 1981. The “Ai gong wen wu yi” manuscript has not yet
been published.
42
Mother to the People”), “Zigao 子羔,” “Lu bang dahan 魯邦大旱” (“The Great Drought of Lu”),
“Zhong gong 仲弓”, “Xiang bang zhi dao 相邦之道” (“The Way of Ministering a State”), “Ji
Kangzi wen yu Kongzi 季康子問於孔子” (“Ji Kangzi asked Kongzi”), “Junzi wei li 君子為禮”
(“The noble man in the conduct of ritual”), “Dizi wen 弟子問” (“The Disciples Asked”),
“Kongzi jian Ji Huanzi 孔子見季桓子” (“Kongzi had an audience with Ji Huanzi”), “Zi dao e 子
道餓” (“The Master spoke of starvation”), and “Yan Yuan wen yu Kongzi 顏淵問於孔子”
(“Yan Yuan asked Kongzi”) texts; the Mawangdui Yi 易 texts listed above; the list of Kongzi-related anecdote titles in the “Rujia zhe yan 儒家者言” (“Sayings of the Ru”) manuscript from the
Fuyang 阜陽 find; and the single fragmentary Kongzi quotation in the Fuyang “Chunqiu shiyu
春秋事語” (“Stories from the Spring and Autumn period”) manuscript.103 One additional Lunyu
19/17–18 fragment was discovered in the late Western Han Xuanquan 懸泉 find.104
● Collectanea. Encyclopedia collections from later periods, e.g., the Yiwen leiju 藝文類聚 (Categorized Collection of Arts and Letters) and Taiping yulan 太平御覽 (Imperial Survey of the
Taiping Era), preserve a number of Kongzi sayings that may have originated in the early period.
Many of these fragments were included by Sun Xingyan in the Kongzi jiyu.
How are Kongzi sayings used in early texts?
There can be no simple answer to this question given the sheer size of the corpus of Kongzi sayings. For a small minority of texts—the collections of Kongzi material listed above—the compilation of Kongzi material was apparently an end in and of itself. But the Kongzi quotations in the
103.See Han Ziqiang 2004, p. 162.
104.Dunhuang Xuanquan Han jian shicui p. 175.
43
vast majority of early texts were instrumental: they were used for various purposes. Understanding those purposes is the first step towards understanding the form and content of Kongzi sayings
and, ultimately, Kongzi’s appeal as a source of quotable wisdom.
When Sun Xingyan compiled the Kongzi jiyu he grouped Kongzi sayings into fourteen categories based for the most part on common themes: “Quan xue 勸學” (“Encouraging learning”),
“Xiao ben 孝本” (“Filiality is the root”); “Wu xing 五性” (“The Five Dispositions”),105 “Liu yi
六藝” (“The Six Arts), “Zhu de 主德” (“Virtue is chief”), “Chen shu 臣術” (“The Art of the
Minister”), “Jiao dao 交道” (“Associating in the Way”), “Lun ren 論人” (“Assessing others”),
“Lun zheng 論政” (“Assessing government”), “Bo wu 博物” (“Knowledgeable about things”),
“Shi pu 事譜” (“Biography”), “Za shi 雜事” (“Assorted matters”), “Yi chen 遺讖” (“Remnant
prognostication texts”), and “Yu yan 寓言” (“Imputed words”). Sun Xingyan’s preference for
thematic typology suggests that he thought of Kongzi sayings primarily as teachings, as sources
of Kongzi’s thought. But Sun Xingyan did not read all Kongzi sayings in this way. Judging from
the “Lun ren” and “Lun zheng” headings, he also thought of Kongzi sayings as performing certain functions or demonstrating certain skills. Still other Kongzi material, e.g., the entries of the
“Shi pu” chapter, was deemed valuable primarily for its historical value, while the “Yu yan” entries were apparently not very valuable at all.106
The Kongzi jiyu’s mixed typology, though helpful, can be improved upon. Whereas Sun
Xingyan thought of Kongzi sayings primarily as teachings, I have found it more useful to think
105.The wu xing 五性have been defined in various ways, but the content of this chapter suggests that Sun
Xingyan’s list paralleled the definition found in the Baihu tongyi (SBCK 8/1a): “What are the ‘five dispositions’?
Humaneness, rightness, ritual propriety, wisdom, and faithfulness” (五常（性）者何？謂仁、義、禮、智、信
也。).
106.The earliest extant text to use the phrase yu yan 寓言 is Zhuangzi book 27 (SBCK 9/11b). For a discussion, see
pp. 104–108.
44
of them primarily in terms of their functions, as texts that do things. As we will see below, more
than a few Kongzi sayings exhibit very little in the way of content, let alone a fully developed set
of teachings or philosophy. But even these sayings have a function that can be described and analyzed, and these functions are crucial to understanding why early authors were so interested in
quoting Kongzi. The categories outlined below do not begin to encompass all the uses of Kongzi
sayings that one finds in received and excavated texts, nor do my examples represent anything
more than a tiny fraction of the available material. They are only meant to illustrate what I believe to be the most common features of the corpus, and to provide a framework for further
study.
○ Dependent Kongzi sayings, p. 47
– Kongzi comments, p. 47
• Comments on anecdotes, p. 48
• Comments on other sayings, p. 53
• Comments on people, p. 59
• Comments on natural phenomena, p. 67
– Kongzi commentaries, p. 68
• Amplifying commentaries, p. 71
• “Greater meaning” (da yi 大義) commentaries, p. 74
• “Clarifying” (ming 明) commentaries, p. 78
• Word and character glosses, p. 81
• Meta-commentaries, p. 82
○ Independent Kongzi sayings, p. 83
– Objects of commentary, p. 87
– Objects of criticism, p. 94
– Proof texts, p. 98
– Records of Kongzi’s life and thought, p. 101
○ “Imputed words” (yu yan 寓言), p. 104
The largest division in my scheme is between “dependent” and “independent” Kongzi sayings.
Dependent sayings respond to other texts; without a prompt text, a dependent Kongzi saying
would not exist. Independent sayings, on the other hand, do not respond directly to other texts
45
and so are more likely to give the impression of being teachings or records of Kongzi’s life and
thought.
Although a useful tool for classifying and analyzing Kongzi sayings, my distinction between
dependent and independent sayings, or between any other categories listed above, is not hard and
fast. As the author of the “Tan gong 檀弓” (“Sandalwood bow”) chapter of the Liji observed,
Kongzi sayings which were properly understood as dependent responses to specific contexts
could circulate as if they were independent teachings:
Youzi asked Zengzi, “Did you ever ask the Master about losing one’s position?” Zengzi said,
“I learned that ‘losing one’s position one should wish for swift poverty; losing one’s life one
should wish for swift decay.’” Youzi said, “This is not the saying of a noble man.” Zengzi
said, “I heard it myself from the Master.” Youzi again said, “This is not the saying of a noble
man.” Zengzi said, “I was with Ziyou when I heard it.” Youzi said, “In that case, there was
something the Master was saying it about.”
When Zengzi reported what Youzi had said to Ziyou, Ziyou said, “Astounding! Youzi’s
words resemble the Master’s.107 Previously the Master was dwelling in Song when he saw that
Master of Horses Huan had a stone coffin made for himself which took longer than three years
to finish. The Master said, ‘What a waste! When he dies, the quicker he decays the better.’
‘Losing one’s life one should wish for swift decay’ was said about Master of Horses Huan.
Whenever Nangong Jingshu returned home he would always bring valuables to court. The
Master said, ‘What wealth! Once out of office, the quicker he becomes poor the better.’
‘Losing one’s position one should wish for swift poverty’ was said about Nangong Jingshu.”
When Zengzi reported what Ziyou had said to Youzi, Youzi said, “Of course. I told you
before that this was not the Master’s saying.” Zengzi said, “How did you know?” Youzi said,
“The Master enacted a regulation in Zhongdu to the effect that the inner coffin should be four
inches thick and the outer coffin five, and this is how I knew that he did not wish for swift
decay. Previously, when the Master lost his position as Minister of Justice in Lu and was about
to go to Chu, he first sent Zixia and then Ranyou ahead of him. This is how I knew that he did
not wish for swift poverty.”
有子問於曾子曰：問喪於夫子乎？曰：聞之矣。喪欲速貧，死欲速朽。有子曰：是非君
子之言也。曾子曰：參也聞諸夫子也。有子又曰：是非君子之言也。曾子曰：參也與子
游聞之。有子曰：然，然則夫子有為言之也。
曾子以斯言告於子游，子游曰：甚哉！有子之言似夫子也。昔者夫子居於宋，見桓
司馬自為石槨，三年而不成。夫子曰：若是其靡也，死不如速朽之愈也。死之欲速朽，
為桓司馬言之也。南宮敬叔反，必載寶而朝。夫子曰：若是其貨也，喪不如速貧之愈
也。喪之欲速貧，為敬叔言之也。
107.This Youzi is probably the disciple identified as You Ruo 有若 in the “Zhongni dizi liezhuan” (Shiji 67.2215),
whose name translates as “having a resemblance.” Thus, it is entirely possible that the use of the word si 似
(resemble) is a play on Youzi’s name. For my discussion of the significance of puns on disciple names, see pp. 111–
115.
46
曾子以子游之言告於有子。有子曰：然。吾固曰：非夫子之言也。曾子曰：
子何以知之？有子曰：夫子制於中都，四寸之棺，五寸之槨，以斯知不欲速朽
也。昔者夫子失魯司寇，將之荊，蓋先之以子夏，又申之以冉有，以斯知不欲
速貧也。108
Youzi’s critique of Zengzi’s Kongzi quotation is a warning about the dangers of decontextualizing Kongzi sayings and treating them a generalized teachings. Deprived of its original frame of
reference, a dependent Kongzi saying might cease to be worthy of a junzi 君子 (noble man).
Considered in light of the present exercise, the episode is also a useful reminder that Kongzi quotation was a fluid practice, and that a Kongzi saying could be interpreted very differently depending on the context of its use. To label a given Kongzi saying as “dependent” or “independent,”
then, is simply to describe its use in the text in which it appears. Unlike Youzi, who was intent
on defending Kongzi’s reputation against inappropriate sayings, I do not mean these labels to imply anything about the original or authentic form of the material.
Dependent Kongzi sayings
● Kongzi comments
“Kongzi comments” are single-serving commentaries that respond to short, discrete texts like
anecdotes and other sayings. Eric Henry has observed of Zhongni yue comments in the Zuozhuan
that they are “brief and readily separable from the main text,” and this observation holds for
Kongzi comments generally.109 I reserve the term “Kongzi commentary” for Kongzi yue sayings
108.Liji (SBCK 2/15a; Kongzi jiyu jiaobu 521–22). The Kongzi jiayu also contains a parallel account of Kongzi’s
doings in Song (SBCK 10/1a). For my argument that the Kongzi saying in question was composed for this specific
anecdote, see p. 94 below.
109.Henry 1999, p. 126. See also Schaberg 2001, p. 179, on Kongzi comments in the Zuozhuan: “in his comments
47
attached to canonical texts, especially the the Shi 詩, Shu 書, and Yi 易. By virtue of their status
and circulation in the early period, canonical texts prompted Kongzi sayings that were longer and
more systematically organized than Kongzi comments on other kinds of material.
○ Comments on anecdotes
The following Kongzi comment from the Lüshi chunqiu (with a parallel version in the
Huainanzi) is typical of the sub-genre:
A certain Ci Fei from Chu acquired a prized sword at Gansui. On his return he was crossing
the Yangzi when two flood-dragons surrounded his boat in the middle of the river. Ci Fei said
to the boatman, “In the past when you saw two flood-dragons surrounding your boat, did you
manage to survive?” The boatman said, “I have never seen such a thing before.” Ci Fei then
pulled up his sleeves and robes and drew his sword, saying, “This [body] is [but] a rotting bag
of bones in a river. How could I care so much about it as to abandon my sword and save
myself?” He then dove into the river and stabbed the dragons, and after killing them he got
back into the boat. Everyone in the boat survived. The king of Chu learned of this and
enfeoffed Ci Fei with the title “wielder of the jade scepter.” Kongzi learned of this and said,
“Excellent! Not ‘abandoning his sword for his own rotting bag of bones’—may this be said of
Ci Fei!”
荊有次非者，得寶劍于干遂，還反涉江，至於中流，有兩蛟夾繞其船。次非謂舟人曰：
子嘗見兩蛟繞船能兩活者乎？船人曰：未之見也。次非攘臂袪衣拔寶劍曰：此江中之腐
肉朽骨也。棄劍以全己，余奚愛焉！於是赴江刺蛟，殺之而復上船，舟中之人皆得活。
荊王聞之，仕之執圭。孔子聞之曰：夫善哉！不以腐肉朽骨而棄劍者，其次非之謂
乎！110
As is common with anecdotes capped with Kongzi wen zhi yue 孔子聞之曰 (Kongzi heard or
learned of this and said), there is no obvious connection between Kongzi and the figures mentioned in the anecdote, nor is there any effort to explain how the Ci Fei story survived the realities of interstate communication in the late Spring and Autumn era (771–fifth century BCE) to
[Kongzi] is normally as uninvolved in the events in question as the anonymous junzi.”
110.Lüshi chunqiu 20/3.2 (SBCK 20/6b–7a; KZJY 9.48/p. 190). For the Huainanzi version, see Huainanzi 12 (SBCK
12/17b; KZJY 9.49/pp. 190–1).
48
reach Kongzi’s ears. For the compilers of these anecdotes, historical realism was beside the
point. What mattered was the judgment itself.111
The Kongzi comment in question introduces a number of elements typical of Kongzi comments generally, beginning with an explicit statement of approval—here shan zai 善哉 (“Excellent!”). Shan zai also figures prominently in exchanges between Warring States persuaders and
rulers and typically expresses the approval of a superior for an inferior.112 In the Ci Fei anecdote,
the juxtaposition of Kongzi’s praise with that of a Chu king implies a hierarchy of judgment in
which the rewards bestowed by temporal political authorities are trumped by Kongzi’s more enduring praise. By giving Kongzi the last word, the Lüshi chunqiu compilers seem to have indicated that his opinion mattered more than that of a long-dead king. But not all early authors who
quoted Kongzi upheld Kongzi as the ultimate authority. The Huainanzi compilers in the early
Western Han capped their version of the Ci Fei story with a Kongzi comment followed by a quotation from the Laozi, a choice consistent with the Huainanzi’s privileging of Laozi over Kongzi:
“Thus Laozi said, ‘It is better to act out of a disregard for life than to value life’” (故老子曰：“
夫唯無以生為者，是賢於貴生焉。).113
The nominalizing particle zhe 者 and the verb wei 謂 (“to refer to”) are also typical of
Kongzi comments and evince an interest in how anecdotes and their characters are labeled.
111.Schaberg 2001, p. 179: “Marked judgment always has about it a temporal ambiguity. One does not know if the
speaker was a contemporary of the judged events or lived in some later time. His authority may derive from autopsy
or from the high ground of retrospection.” At least one Eastern Han author reflected the textual culture of his era
when he reworked what was originally a Zhongni yue saying from the Zuozhuan with the words “Kongzi read this
record and said” (孔子讀其志曰). See Kongzi jiayu 41 (SBCK 9/19a).
112.See, e.g., the dialogue between Mengzi and King Xuan of Qi 齊宣王 at Mengzi 1B/5.
113.In the Huainanzi chapter in which this anecdote appears, “Ying dao 應道” (“Responding to the Way”), every
anecdote is similarly capped with an apposite Laozi quotation. No Huainanzi chapter privileges Kongzi sayings in
the same way.
49
Kongzi comments do not simply evaluate the various situations and figures of Warring States
anecdotal literature. They also call attention to the act of predication itself in a way that hints at
the pedagogical motivations behind such comments.114 Note the use of the particle zhe 者 in the
following Kongzi comment from the Han Feizi (with a parallel in the Lüshi chunqiu):
Duke Wen of Jin attacked Yuan. After packing provisions for ten days, he promised his
grandees a ten-day campaign. On the tenth day he arrived at Yuan but the capital had still not
fallen, so he announced a retreat and called an end to the fighting. Some men who had come
out of Yuan said, “Yuan will fall in three days,” and so Duke Wen’s subjects and advisors
remonstrated with him, saying, “Yuan’s food supplies have been used up and its force is spent.
My Lord should wait them out.” Duke Wen said, “I promised the men a ten-day campaign. If I
do not leave now, I will lose their trust. Winning Yuan and losing their trust—this I will not
do.” Thereupon he stopped the fighting and left. The people of Yuan heard of this and said,
“How can we not go to a lord as trustworthy as he?” They then submitted to Duke Wen. The
people of Wei heard this and said, “How can we not follow a lord as trustworthy as he?” They,
too, submitted to Duke Wen. Kongzi learned of this and memorialized it, saying, “‘Attacking
Yuan and winning Wei’ [depends on] trustworthiness.”
晉文公攻原，裹十日糧，遂與大夫期十日。至原十日而原不下，擊金而退，罷兵而去。
士有從原中出者曰：原三日即下矣！群臣左右諫曰：夫原之食竭力盡矣，君姑待之。公
曰：吾與士期十日，不去，是亡吾信也。得原失信，吾不為也。遂罷兵而去。原人聞
曰：有君如彼其信也，可無歸乎！乃降公。衛人聞曰：有君如彼其信也，可無從乎！乃
降公。孔子聞而記之曰：攻原得衛者，信也。115
Here, too, zhe 者 highlights a specific topic or expression—the four-word phrase “attacking
Yuan and winning Wei” (攻原得衛)—and tags it with a value—xin 信 (trustworthiness)—explicitly endorsed in the story. (As it happens, the Lüshi chunqiu version of the anecdote preserves
the same comment but introduces it with yue instead of Kongzi yue and omits any mention of
Kongzi.116 Why Kongzi was felt to be an appropriate mouthpiece for the one comment but not
the other is an open question.117)
114.Compare the Kongzi comment on an anecdote featuring Master Cheng of Hou at Lüshi chunqiu 20/8, “Guan
biao 觀表” (SBCK 20/21a; KZJY 9.52/p. 192; translation adapted from Knoblock & Riegel 2000, p. 341): “When
Kongzi learned of this he said, ‘[The saying about how] the wise can be told secrets and the humane can be
entrusted with goods surely refers to Viscount Cheng of Hou” (孔子聞之曰：夫智可以微謀、仁可以託財者，其
郈成子之謂乎！).
115.Han Feizi 32 (SBCK 11/10b; KZJY 6.34/p. 132).
116.Lüshi chunqiu 19/6 (SBCK 19/16a).
117.This disparity might indicate that Kongzi himself was not felt to be an integral part of the Duke Wen anecdote
50
A curious feature of these Kongzi comments is how little they contribute to our understanding of the events in question. Notice that the bulk of the Ci Fei comment is taken directly from
the anecdote itself:
<anecdote>Ci Fei then pulled up his sleeves and robes and drew his sword, saying, “This
[body] is [but] a rotting bag of bones in a river. Why should I care so much about it as to
abandon my sword and save myself?”
次非攘臂袪衣拔寶劍曰：此江中之腐肉朽骨也。棄劍以全己，余奚愛焉！
<comment>Kongzi learned of this and said, “Excellent! ‘Not abandoning his sword for his
own rotting bag of bones’—may this be said of Ci Fei!”
孔子聞之曰：夫善哉！不以腐肉朽骨而棄劍者，其次非之謂乎？
Likewise, the Duke Wen comment simply reiterates a value—xin 信 (“trustworthiness”)—that
Duke Wen had already articulated within the anecdote itself. The derivative nature of the Kongzi
comments is all the more striking when contrasted with the Laozi yue comment in the Huainanzi
Ci Fei anecdote (“Thus Laozi said, ‘It is better to act out of a disregard for life than to value
life.’”), which helpfully pinpoints its moral with an apt quotation which does not merely copy Ci
Fei’s words. If explanation was not the goal of these Kongzi comments, then what was their
value?
The answer to that question lies in the form of the quotation marker in the Han Feizi passage:
“Kongzi learned of this and memorialized it, saying” (孔子聞而記之曰). In most contexts, ji 記
translates simply as “to note down” or “record.”118 But in this instance ji seems to refer more
generally to the act of memorializing or commemorating a given situation or figure, here with an
epitomizing phrase or idea. This usage of ji appears in another Kongzi saying from the Lüshi
tradition. On this point see Schaberg 2005, p. 19: “the commonplaceness of many of the cited judgments [shows]
that these had value first as words, and only secondarily as words associated with a particular individual.”
118.The Kongzi jiyu includes a number of Kongzi yue sayings that begin or end with the phrase dizi ji zhi 弟子記之
(“Disciples take note”). See KZJY 2.6/p. 16, 10.54/p. 231, and 10.57/p. 232. Other formulations include shi 識之 and
mian zhi 勉之.
51
chunqiu that follows an anecdote in which Kongzi wrongly suspects his disciple Yan Hui of having stolen some food:
Confucius sighed and said, “What I believed was my eyes, but it appears that my eyes cannot
be trusted; what I depended on was my mind, but it appears that my mind cannot be depended
upon. Disciples, take note: knowing others is assuredly not easy.”
孔子歎曰：所信者目也，而目猶不可信；所恃者心也，而心猶不足恃。弟子記之：知人
固不易矣。119
Notice that Kongzi only commands his disciples to “take note of” the final six characters of the
thirty-two-character saying. Translating ji as “record’ would imply that the twenty-six preceding
characters were somehow less important or less worthy of remembrance than the final six. But
the point of ji is rather that Kongzi’s disciples (and, by extension, the audience of the anecdote)
should remember the final six characters as a handy tagline or catch phrase for the entire episode.
Loosely translated, Kongzi is telling his followers, “File this under ‘knowing others is assuredly
not easy.’” The comments on Ci Fei and Duke Wen perform a similar function, singling out core
phrases (“Not abandoning his sword for his own rotting bag of bones”) or terms (xin 信) to stand
as verbal insignia for these figures’ exploits. Far from advancing any particular intellectual agenda, the comments seem to anticipate that such stories would have a life beyond the texts, and that
audiences required some guidance in order to successfully process, remember, and quote these
stories in new settings. In this way, they occupy the gap between understanding an anecdote and
skillfully deploying that understanding in new contexts.120
119.Lüshi chunqiu 17/3 (SBCK 17/9b; KZJY 13.32/p. 289. Translation adapted from Knoblock & Reigel 2000, p.
418. For additional examples, including exhortations to “note” (zhi zhi 志之/shi zhi 識之) or “work on” (mian zhi 勉
之), see Appendix 1:
120.Schaberg 2005, p. 2: “written anecdotal accounts, whether in the Zuozhuan or other works, suggest the
usefulness of historical narration in elite conversation, persuasion, and commemoration. Written lore, wherever we
find it in early China, bespeaks a widespread passion for substantiated judgment. This was not a devotion to
historical knowledge itself; the problem of verification was seldom thematized. Instead it was a reflex of a rhetorical
habit of adducing images of the past in support of present polemical needs.”
52
The prevalence of the idea that “knowing others” (zhi ren 知人) is “not easy” (bu yi 不易) or
“difficult” (nan 難) suggests another motivation for Kongzi’s injunction to “take note” of his
words: anchoring free-floating, widely circulating expressions to specific contexts.121 Far from
originating the observation that knowing others is difficult, the Kongzi of the Lüshi chunqiu
comment perhaps deserves credit only for identifying a concrete instance of a widely circulating
proverb. As we will see below, such proverbs frequently provided the raw materials for Kongzi
sayings.
○ Comments on other sayings
Considered from the perspective of comments on anecdotes, the use of Kongzi sayings to comment on other sayings reflects a more general interest in controlling the interpretation of received
textual traditions. Yet whereas anecdotes circulated as self-contained textual units to which
Kongzi yue and other kinds of comments accrued, the sayings spotlighted by Kongzi comments
were treated much more flexibly as wisdom which could be embedded, amplified, or edited as an
author saw fit. Often that wisdom originated in canonical traditions like the Shi, as in these
examples from the Lüshi chunqiu and Hanshi waizhuan:
<Lüshi chunqiu>A Shi says, “Holding the reigns like spinning silk.” Kongzi said, “Examine
this saying carefully—it can be used to rule the world.”
詩曰：執轡如組。孔子曰：審此言也可以為天下。
<Hanshi waizhuan>Kongzi understood that the Way was easy to follow, saying, “The Shi say,
‘Guiding the people is very easy.’ These are not empty words.”
121.See, e.g., Kongzi jiayu 12 (SBCK 3/5a), Da Dai liji 60 (SBCK 6/4a; KZJY 9.16/p. 174), and Shiji 79.2416. The
locus classicus may be the “Gao You mou 皋陶謀” (“Counsels of Gao Yao”) chapter of the Shangshu (SBCK 2/7a).
53
孔子知道之易行，曰：詩云：牖民孔易。非虛辭也。122
Note the shift in terminology the Lüshi chunqiu example from labeling the four-character line a
Shi to labeling it a yan 言 (saying). The two Shi in which this line appears, “Jian xi 簡兮” (#38,
“So Grand”) and “Da shu yu tian 大叔于田” (#78, “Shu in the Hunting-Fields”),123 contain any
number of four-character lines, but only this line qualifies as (a) a yan and (b) advice on universal rule. Here as elsewhere, Kongzi sayings highlight and then transform potent ideas and phrases into objects of contemplation and quotation.
Sometimes Kongzi comments responded to proverbs of unknown origin, as in this example
from the Lüshi chunqiu (with parallels in the Shuiyuan and Kongzi jiayu):
When Kongzi had an audience with Duke Ai of Lu, the Duke said, “Someone once said to me,
‘One who would rule a state need only rule from the audience chamber.’ I took this to be an
absurd saying.” Kongzi said, “This is not an absurd saying. I have heard that ‘those who
obtain it in themselves obtain it in others; those who lose it in themselves lose it in others.’
Only someone who understands reverting to himself can bring order to the world without
leaving his gates.”
孔子見魯哀公，哀公曰：有語寡人曰：為國家者，為之堂上而已矣。寡人以為迂言也。
孔子曰：此非迂言也。丘聞之：得之於身者得之人，失之於身者失之人。不出於門戶而
天下治者，其唯知反於己身者乎！124
This is the second example considered thus far (the first was on p. 52) in which a Kongzi saying
is part of an anecdote in which Kongzi also appears as a character. But the dialogical framework
should not distract us from noticing that the core Kongzi saying essentially functions as a comment on another saying. Kongzi’s explicit evaluation comes at the very beginning of the comment—“This is not an absurd saying” (ci fei yu yan 此非迂言)—just as shan zai 善哉 prefaced
the comment on Ci Fei. It then introduces a second piece of received wisdom in order to clarify
122.Hanshi waizhuan 5 (SBCK 5/10b) and Lüshi chunqiu 3/3 (SBCK 3/7b; KZJY 5.4/p. 69).
123.SBCK 2/13b & 10/4a, respectively.
124.Lüshi chunqiu 3/3 (SBCK 3/7b; KZJY 6.40/p. 134). See also Kongzi jiayu 13 (SBCK 3/16b) and Shuiyuan
(SBCK 7/3a; KZJY 6.41/p. 134).
54
the evaluation of the first, and it concludes with a pithy summary which cements the connection
between the original saying and what Kongzi had previously “heard” (wen 聞). Here, too, we
find that the added value of a Kongzi comment lies in the connections it makes among various
ideas, but not necessarily the substance of the ideas themselves.
Although the following Kongzi comments from the Zuozhuan might be categorized as comments on anecdotes insofar as they are attached to narratives, they are also framed as responses
to other sayings of anonymous origin:125
Zhongni said, “There is an ancient motto that says, ‘Suppressing oneself and returning to ritual
is humane.’ How truly excellent! If King Ling of Chu had been capable of this, how could he
have been shamed at Ganxi?”
仲尼曰：古也有志：克己復禮，仁也。信善哉！楚靈王若能如是，豈其辱於乾谿。126
Zhongni said, “There is a motto that says, ‘the words are to be adequate to what is on the
person’s mind, and the patterning is to be adequate to the words.’ If a person does not use
language, who will know his aims? And if the language lacks patterning, it will not go far.
When Jin was hegemon, Zheng’s invasion of Chen would not have been successful had its
words had not been patterned. Heed your words!”
仲尼曰：志有之：言以足志，文以足言。不言誰知其志？言之無文，行而不遠。晉為
伯，鄭入陳。非文辭不為功。慎辭哉！127
Both Zhongni yue sayings mark the quoted material as zhi 志, a word which generally refers to a
person’s state of mind or intent but which is also glossed as ji 記 by a number of early commentators.128 In this latter sense, zhi designates something “noted down,” a written record or some125.See Henry 1999 and Schaberg 2001, p. 397, n. 62, for a list of Kongzi comments in the Zuozhuan.
126.Zuozhuan Zhao 12 (SBCK 22/15b; KZJY p. 418). It is not entirely clear what the “shame at Ganxi” refers to
here. Legge (1991, p. 649) notes a curious inconsistency between the Chunqiu and Zuozhuan accounts of King
Ling’s death. While the Chunqiu chronicle for the following year (Zhao 13, SBCK ) states that Duke Ling was killed
at Ganxi by a certain Gongzi Bi, the Zuozhuan records that the king’s army was persuaded to switch sides at Ganxi
and that the king committed suicide at another location. Du Yu 杜預 (222–285) attempts to square this contradiction
by arguing that Ganxi marked the beginning of King Ling’s downfall (SBCK 23/1a). However, the Zuozhuan
account of King Ling’s end begins earlier than the events at Ganxi. It is possible that the Zhongni yue comment’s
seeming agreement with the Chunqiu account of King Ling’s demise indicates that its author was more familiar with
the Chunqiu version than with the Zuozhuan version.
127.Zuozhuan Xiang 25 (SBCK 17/14b; KZJY p. 415).
128.See, e.g., a Zheng Xuan gloss preserved in the Zhouli zhengyi 周禮正義 (SBCK 6/47b).
55
thing that is worthy of being noted down or remembered, hence “motto.”129 In the first comment,
which follows an anecdote about the failings of King Ling of Chu 楚靈王 (r. 540–529 BCE), it
is the quoted zhi and not any person or action which elicits Kongzi’s explicit approval (shan zai
善哉). Only in the second half of the comment is the motto applied to King Ling, whose “shame
at Ganxi” and ultimate death are explained by his failure to grasp the motto’s wisdom. As a
result, “suppressing oneself and returning to ritual is humane” is presented as much more than a
memorable saying. These are words to live by...or else.
The second Kongzi comment follows an anecdote about Zichan 子產, a famous minister of
Zheng 鄭, and resembles the first insofar as it cites an anonymous saying which it applies to the
situation at hand. But the comment does much more with its source material, refashioning it into
a series of corollary aphorisms. The rhetorical question which follows the original saying—“If
one does not speak, who will know his aims?” (不言誰知其志)—rephrases the first four characters of the source text, and the two parallel tetrametric phrases—“If a person does not use language, who will know his aims?” (言之無文，行而不遠)—rephrase the second four characters.
And for anyone who fails to grasp the gist of the motto, the comment concludes with a pithy
warning: “Heed your words!” (shen ci zai 慎辭哉). One might be tempted to treat such blatantly
redundant amplification as mere window dressing for the comment’s core concern—praising the
rhetorical skills of Zichan. But the fact that it devotes more space to the motto’s amplification
than to its application suggests that “patterned” (wen 文) elaboration was precisely the point: it
129.The history of the word “motto” in Romance languages makes it a particularly apt translation. According to the
Oxford English Dictionary online (http://oed.com; accessed 2/21/11), a “motto” is “a word, sentence, or phrase
attached to an impresa or emblematical design to explain or emphasize its significance. Later also: a short sentence
or phrase inscribed on an object, expressing a reflection or sentiment considered appropriate to its purpose or
destination; a maxim or saying adopted by a person, family, institution, etc., expressing a rule of conduct or
philosophy of life.”
56
demonstrates the motto’s wisdom and rhetorical potential by making it “go far” (yuan 遠). Given
this emphasis, one wonders whether the comment was designed to illuminate the story of King
Ling, or whether it was attached to the King Ling story in order to illuminate the motto.
I have encountered no better example of the use of a Kongzi comment to create a memorable
saying from pre-existing material than this episode from the “Gui gong 貴公” (“Esteeming impartiality”) chapter of the Lüshi chunqiu:
A man of Chu lost his bow and had no desire to look for it, saying, “A man of Chu lost it and a
man of Chu will find it, so why look for it?” Kongzi heard this and said, “Get rid of ‘of Chu’
and it’ll be fine.” Lao Dan heard this and said, “Get rid of ‘a man’ and it’ll be fine.” Thus it
was Lao Dan who was perfectly impartial.
荊人有遺弓者，而不肯索，曰：荊人遺之，荊人得之，又何索焉？孔子聞之曰：去其荊
而可矣。老聃聞之曰：去其人而可矣。故老聃則至公矣。130
The anecdote presents itself as a statement on the difference between Kongzi and Laozi. But its
author frames that difference not in terms of their worldviews or philosophies but in terms of
rhetoric, treating Kongzi and Laozi sayings as two different strategies for enhancing the profundity and memorability of pre-existing material. Consider the progression from the anonymous
man of Chu’s saying to the Kongzi and Laozi versions:
A man of Chu lost it and a man of Chu will find it, so why look for it?
荊人遺之，荊人得之，又何索焉？
A man lost it and a man will find it, so why look for it?
人遺之，人得之，又何索焉？
It was lost and it will be found, so why look for it?
遺之，得之，又何索焉？
The value prized in the anecdote is gong 公 (“impartiality”). The Kongzi version performs gong
by eliminating words to create a shorter and more abstract saying, and Laozi shows himself to be
“perfectly impartial” (zhi gong 至公), i.e., to have the broadest possible perspective, by applying
130.Lüshi chunqiu 1/4 (SBCK 1/9b–10a; KZJY 17.30/p. 405).
57
the same algorithm to Kongzi’s version. (Tellingly, the Lüshi chunqiu’s praise of Laozi was not
tolerated by Liu Xiang 劉向 [79–8 BCE], who in the Shuiyuan omitted the Laozi saying altogether and instead concluded the episode with a description of Kongzi as “greatly impartial” [da
gong 大公].)131 How the intended audience of the Lüshi chunqiu received this anecdote is an
open question. But my hunch is that it might have read as a recipe for generating maximally
pithy and memorable language à la Laozi or Kongzi.
Another example from the Hanshi waizhuan presents a Kongzi comment as part of a dialogue between Kongzi and three of his closest disciples:
Zilu said, “When a person is good to me, I am then good to him. When a person is not good to
me, I am not good to him.” Zigong said, “When a person is good to me, I am then good to
him. When a person is not good to me, I will simply draw him in and advance or retreat [as
necessary].” Yan Hui said, “When a person is good to me, I am then good to him. When a
person is not good to me, I am still good to him.” The three followers each differed with
respect to what they upheld, and so they consulted Kongzi. The Master said, “What You
[=Zilu] says is a saying for Man and Mai [barbarians]; what Ci [=Zigong] says is a saying for
friends; what Hui says is a saying for kin.”
子路曰：人善我，我亦善之；人不善我，我不善之。子貢曰：人善我，我亦善之；人不
善我，我則引之進退而己耳。顏回曰：人善我，我亦善之；人不善我，我亦善之。三子
所持各異，問於夫子。夫子曰：由之所持，蠻貊之言也；賜之所言，朋友之言也；回之
所言，親屬之言也。132
As with the “man of Chu” example from the Lüshi chunqiu, the contrived nature of the episode is
evident in the overlap among the three sayings, which are identical save for a few variations:
<Zilu>When a person is good to me, I am then good to him. When a person is not good to me,
I am not good to him.
人善我，我亦善之；人不善我，我不善之。
<Zigong>When a person is good to me, I am then good to him. When a person is not good to
me, I will simply draw him in and advance or retreat [as necessary].
人善我，我亦善之；人不善我，我則引之進退而己耳。
<Yan Hui>When a person is good to me, I am then good to him. When a person is not good to
me, I am still good to him.
人善我，我亦善之；人不善我，我亦善之。
131.Shuiyuan book 14 (SBCK 14/6b; KZJY 17.31/p. 405).
132.Hanshi waizhuan book 9 (SBCK 9/4a; KZJY 8.2/p. 159). Here I have omitted the capping Shi quotation.
58
The Kongzi comment’s strategy is to point out the disciple sayings’ limitations as abstract principles, with each saying being appropriate only to a specific circumstance. Whereas the Lüshi
chunqiu Kongzi comment performed gong by broadening the wisdom inherent in the earlier saying, here the comment’s gong-ness is manifest in the breadth of Kongzi’s perspective. Just as important as knowing how to craft memorable sayings was recognizing the situations in which
those sayings could be appropriately deployed.
○ Comments on people
From comments on anecdotes and sayings it is but a small step to comments on people,133 one of
the most frequently encountered kinds of Kongzi saying. Sun Xingyan devoted an entire chapter
of the Kongzi jiyu—“Lun ren” 論人 (“Evaluating others”)—to this subgenre, whose most common marker is the question he ren 何人 or he ru ren 何如人 (“What sort of man is X”).
To reiterate a point from the introduction to this section, whether we label a given Kongzi
saying as a comment on a person or anecdote or saying matters relatively little. As we saw
above, Kongzi’s comments on the Ci Fei anecdote from the Lüshi chunqiu and the King Ling
story from the Zuozhuan were also comments on Ci Fei and King Ling themselves. What distinguishes Kongzi comments on people as a distinct category of Kongzi sayings is an interest in the
problems and possibilities of character evaluation itself, one of the core functions of Kongzi quotation practice in the early period.
133.The connection between famous sayings and the conduct of historical figures is also emphasized by this
aphorism from the Xiang 象 commentary to the “Greater Accumulation” (“Da xu 大畜”) hexagram of the Zhou
Changes: “A noble man is familiar with many former sayings and past actions, and thereby accumulates his virtue”
(君子以多識前言往行，以畜其德。). See SBCK 3/6b.
59
One gets an idea of the diversity of this category of Kongzi sayings from the following
Kongzi yue/Zhongni yue comments in the Zuozhuan:
Kongzi said, “A Shi says, ‘When the people have so many iniquities, do not yourself establish
iniquity.’ This refers to Xie Ye.”
孔子曰：詩云：民之多辟，無自立辟。其洩冶之謂乎。134
Zhongni said, “There were three occasions when Zang Wenzhong displayed a lack of
humaneness and three when he displayed a lack of wisdom. He kept Zhan Qin [=Liuxia Hui
柳下惠] in a lowly position; he removed the Six Gates; and he had his concubines weave rush
mats [for sale]—these are the three instances of his lack of humaneness. He used items above
his rank; he allowed a reverse-order sacrifice to go forward; and he sacrificed to the yuanju [a
mysterious sea-bird]—these are the three instances of his lack of wisdom.
仲尼曰：臧文仲其不仁者三，不知者三：下展禽，廢六關，妾織蒲，三不仁也；作虛
器，縱逆祀，祀爰居，三不知也。135
When Kongzi learned of Zichan’s death, he wept and said, ‘The ancients’ concern [for the
people] was preserved in him.’”
及子產卒，仲尼聞之，出涕曰：古之遺愛也。136
The format of Kongzi’s comment on Xie Ye, a “grandee” (dafu 大夫) executed by Duke
Ling of Chen 陳靈公 (r. 613–599 BCE) after he dared to criticize the Duke’s yin 淫 (“depravity”), should look familiar in light of the comments considered above. Here it is not an anonymous saying or a handy tagline but a Shi couplet which is invoked to commemorate Xie Ye.
In contrast, Kongzi’s comment on Zang Wenzhong shows no interest in quoting other material. In his analysis of Zhongni yue and Kongzi yue comments in the Zuozhuan, Eric Henry pointed to this example to argue that Kongzi comments functioned as “global character assessments”
in contrast to junzi yue 君子曰 (the noble man says) comments’ focus on the rightness or wrongness of specific acts.137 The Kongzi comment on Zang Wenzhong follows an anecdote about the
134.Zuozhuan Xuan 9 (SBCK 10/11a; KZJY p. 414).
135.Zuozhuan Wen 2 (SBCK 8/5a; KZJY p. 413). For the translation, see Henry 1999, p. 138.
136.Zuozhuan Zhao 20 (SBCK 24/12b; KZJY 421).
137.Henry 1999, p. 139. See also Nylan & Wilson 2010, p. 81.
60
rearranging of ancestral tablets in the state temple, a gross violation of ritual propriety which occurred during Zang Wenzhong’s tenure as minister. In Henry’s analysis, the comment argues
against the common perception of Zang Wenzhong as a virtuous man, in this way “alert[ing] the
reader to the possibility of a response different from the one he has already formed; thus [such
comments] serve a corrective or enlarging function in the text.”138 This corrective function appears elsewhere in the Zuozhuan in a Kongzi comment on Zichan: “Considering this, I do not
believe it when people say that Zichan was not humane” (仲尼聞是語也。曰：以是觀之，人
謂子產不仁，吾不信也。).139 Henry’s observations about Zuozhuan Kongzi comments closely
parallel my discussion of the ji 記 function of Kongzi comments in other texts. Such comments
are designed to epitomize historical figures just as other Kongzi comments epitomize individual
anecdotes.140
Kongzi’s eulogy of Zichan, a famous minister of Zheng 鄭, functions in a similar way but
differs from the examples above insofar as it is prefaced by a description of Kongzi’s non-verbal
reaction to the news of Zichan’s death. The brief, two-word statement chu ti 出涕 ([Kongzi]
wept) turns what would otherwise be a simple saying into an emotionally charged mini-scene.
Nonverbal cues like this one were a way to enhance the emotional intensity of Kongzi sayings
and thereby imbue them with personality.141
Kongzi comments on people also extended to legendary figures from the distant past, as in
138.Henry 1999, p. 148. See Lunyu 5/13 and 5/17 for two additional Kongzi criticisms of Zang Wenzhong.
139.Zuozhuan Xiang 31 (SBCK 19/16a; KZJY p. 416).
140.For an explicit ji zhi 記之 exhortation in a Kongzi comment on a person, see Shuiyuan book 7, “Zheng li 政理”
(SBCK 7/5b–6a; KZJY 10.57/pp. 232–33): “Kongzi said, ‘Disciples take note: Duke Huan was a lord-protector, and
Guang Zhong was an accomplished assistant... (孔子曰：弟子記之，桓公，霸君也；管仲，賢佐也；猶有以智
為愚者也...).
141.Kongzi’s “sighs” (tan 歎) are used to similar effect in Kongzi sayings, e.g., at Han Feizi 36 (SBCK 15/2a; KZJY
6.36/pp. 132–33) and in the Yanzi chunqiu (KZJY 7.7/p. 148).
61
this comment on a Zhuangzi anecdote about King Wen, who is portrayed there as having fabricated a prophetic dream in order to persuade his officials to turn the state’s management over to
a lowly fisherman:
Yan Yuan asked Zhongni, “Did King Wen fall short? Why did he use a [fake] dream to get his
way?” Zhongni said, “Quiet! Do not speak. King Wen was perfect, so how can you evaluate
and criticize him? He only used [the dream] to handle that particular situation.”
顏淵問於仲尼曰：文王其猶未邪？又何以夢為乎？仲尼曰：默！汝無言。夫文王盡之
也，而又何論刺焉？彼直以循斯須也。142
The Kongzi saying simply asserts King Wen’s unimpeachability in response to the doubt raised
by Yan Hui. King Wen was “perfect” (jin zhi 盡之), therefore he cannot be criticized. Here we
see that Kongzi comments did not just provide objective evaluations of historical (or quasi-historical) figures. They also policed the stories behind those figures in order to maintain their reputations and symbolic authority. Of course, the fact that this anecdote appears in the Zhuangzi
might indicate that it was meant as a satirical portrayal of Kongzi’s dogmatic support of the ancient sages. However, Sima Qian in book 61 of the Shiji also credits Kongzi with rendering a
similar service to other legendary figures: “Although Boyi and Shuqi were men of accomplishment, they needed the Master to make their names even more conspicuous” (伯夷、叔齊雖賢，
得夫子而名益彰。). For Sima Qian at least, Kongzi deserved credit for preserving the memory
of truly virtuous men, perhaps because Kongzi comments performed precisely this function.
One of the sources Sima Qian drew from in the course of compiling the Shiji was the “Wu di
de 五帝德” (“Virtues of the Five Emperors”) chapter of the Da Dai Liji, which features Kongzi
responding to a series of questions about legendary sage rulers, beginning with the Yellow Emperor (Huangdi 黃帝):
Zai Wo asked Kongzi, “Previously I had heard Rong Yi say that the Yellow Emperor lived for
300 years. May I ask, was the Yellow Emperor a man? Was he not a man? How could he have
142.Zhuangzi 21 (SBCK 7/38a–b; KZJY 17.15/p. 392).
62
lived for three-hundred years?
Kongzi said, “Yu [=Zai Wo]! [The doings of the Xia founder] Yu, [the Shang founder]
Tang, [the Zhou Kings] Wen, Wu, Cheng, and the Duke of Zhou can be thoroughly observed.
The Yellow Emperor is more ancient still—why would you [ask such a question]? Former
masters spoke about him only with difficulty.”
Zai Wo said, “Traditions about high antiquity, abstruse explanations, and discriminations
of affairs long concluded only obscure and confuse [the truth of the matter]. Surely this is not
the way of the noble man. That being so, my question is a natural one.”
宰我問於孔子曰：昔者予聞諸榮伊，言黃帝三百年。請問黃帝者人邪？亦非人邪？何以
至於三百年乎？
孔子曰：予！禹、湯、文、武、成王、周公，可勝觀也！夫黃帝尚矣，女何以為？
先生難言之。
宰我曰：上世之傳，隱微之說，卒業之辨，闇昏忽之，意非君子之道也，則予之問
也固矣。143
In reply, Kongzi offered a summary of the Yellow Emperor’s virtues which concluded with the
following statement:
[The Yellow Emperor] lived and the people reaped the benefits [of his rule] for 100 years.
When he died, the people were in awe of his spirit for [another] 100 years. When that came to
an end, the people used his teachings for [another] 100 years. Thus it is said [that he lived for]
300 years.
生而民得其利百年，死而民畏其神百年，亡而民用其教百年，故曰三百年。
“Wu di de” gives no explanation for its rejection of the Methuselah-esque legend of the Yellow
Emperor aside from what is quoted here. (Interestingly, Kongzi’s rationalizing account of the
Yellow Emperor’s longevity reflects spirit of Lunyu 7/21: “The Master did not talk about oddities, physical strength, chaos, or spiritual matters” [子不語：怪、力、亂、神。].) My suspicion is that the Kongzi saying’s insistence on the Yellow Emperor’s humanity might have been
a way to preserve his status as an approachable exemplar of the traditional virtues, thereby wresting control of the Yellow Emperor legend away from those who would exploit his image for other purposes. To understand the Yellow Emperor’s legacy, the text seems to suggest, we need
only appreciate the potency of his virtue (de 德), not any spiritual or divine (shen 神) qualities he
143.Da Dai Liji 62 (SBCK 7/1a–b; KZJY 6.9/p. 117). For Sima Qian’s use of this text, see concluding comment at
Shiji 1.46.
63
might have possessed.
This is one respect in which Kongzi comments on people differ from other kinds of Kongzi
comments. Many of the historical (or quasi-historical) figures who populated Warring States
anecdotal literature were powerful symbols in their own right who transcended particular anecdote traditions. Some of these figures, including King Wen and even Kongzi himself, were
claimed by particular intellectual lineages and thus acquired a special authority. Maintaining
such figures’ symbolic status was a far more pressing problem than whether this or that saying or
anecdote was worthy of Kongzi’s approval.
Some Kongzi comments thrust Kongzi’s own character front and center as a standard against
which to measure other figures, as in the following episode from the Huainanzi:
Someone asked Kongzi, “What sort of man is Yan Hui?” Kongzi said, “He is a humane man. I
am nothing like him.” “What sort of man is Zigong?” “He is a well-spoken man. I am nothing
like him.” “What sort of man is Zilu?” “He is a courageous man. I am nothing like him.” The
guest said, “These three are all more accomplished than you, Master, and yet they serve you.
Why is that?” Kongzi said, “I can be both humane and merciless, well-spoken and
stammering, courageous and timid. I would not exchange my single Way for the abilities of
these three disciples.” Kongzi understood how to carry out [his Way].
人或問孔子曰：顏回何如人也？曰：仁人也。丘弗如也。子貢何如人也？曰：辯人也。
丘弗如也。子路何如人也？”曰：勇人也。丘弗如也。賓曰：三人皆賢夫子，而為夫子
役。何也？孔子曰：丘能仁且忍，辯且訥，勇且怯。以三子之能，易丘一道，丘弗為
也。孔子知所施之也。144
Here Kongzi is made to evaluate three of his closest and most famous disciples, each of whom
exemplifies certain admirable qualities. But they are all found wanting in comparison with their
master, who embodies all of these qualities in the appropriate measure and thus demonstrates the
gong 公-ness (impartiality) of his perspective. As I will argue in the conclusion to this chapter,
it is tempting to imagine that the perception of Kongzi as a figure of exemplary virtue grew out
144.Huainanzi 18 (SBCK 18/14b; KZJY 3.22/pp. 34–35). These associations are attested in other sources, e.g., in
Lunyu 11/3. For additional discussion, see pp. 112ff.
64
of the use of Kongzi comments to evaluate character, on the assumption that only a truly virtuous
man could accurately assess the virtues of others.145
This perception of Kongzi may also account for the widespread use of Kongzi sayings to
comment on the possibilities and problems of character evaluation itself. One of these sayings is
the Lüshi chunqiu passage discussed above (p. 52) in which Kongzi enjoins his disciples to ji 記
(take note of) the lesson that “knowing others is not easy.” Another is Lunyu 2/10:
The Master said, “Examine what he relies on, observe what he follows, and investigate where
he resides. How could a person hide? How could a person hide?
子曰：視其所以，觀其所由，察其所安。人焉廋哉？人焉廋哉？146
Chapter thirty-two of the Zhuangzi furnishes a more spectacular example of a Kongzi saying
which explores the problems of knowing others:
Kongzi said, “The hearts of others are more treacherous than mountains and rivers, and more
difficult than knowing Heaven. Heaven has the phases of spring and autumn, winter and
summer, dawn and dusk, but other people present an opaque appearance and bury their true
feelings deep.”
孔子曰：凡人心險於山川，難於知天。天猶有春秋冬夏旦暮之期，人者厚貌深情。147
The passage goes on to articulate the “nine proofs” (jiu zheng 九徵) by which a junzi evaluates
others, e.g., “sending him far away to observe his loyalty” (遠使之而觀其忠) or “unexpectedly
questioning him to observe his understanding” (卒然問焉而觀其知).148
Strictly speaking, the Lunyu and Zhuangzi examples count as independent sayings insofar as
they do not respond to other texts or figures. But theoretical statements like these are not entirely
145.In a few texts, Kongzi’s ability to evaluate others was cast in an almost supernatural light, as in a passage from
the Hanshi waizhuan in which Kongzi detects a man’s virtue from the sound of him crying. See Hanshi waizhuan 9
(SBCK 9/2a; KZJY 2.5/pp. 15–16) and its parallel at Shuiyuan 10 (SBCK 10/17b–18a; KZJY 2.6/p. 16).
146.Stephen Owen (1992, pp. 19–22) has pointed to this saying as an example of the early hermeneutics of character
which gave rise to Chinese literary theory. Comments like this one “promis to reveal the complex conditions that
inform human actions and utterances.”
147.Zhuangzi 32 (SBCK 10/19b; KZJY 9.41/p. 188).
148.Mattias Richter (2005) has pointed to this passage as one of several texts of a textual tradition dealing with the
evaluation of official candidates. For additional discussion, see pp. 217–224 below.
65
independent of the practice of using Kongzi sayings to evaluate others. Kongzi may have been
felt to be an appropriate mouthpiece for statements on the hermeneutics of character precisely
because Kongzi yue sayings demonstrated over and over again Kongzi’s skill at evaluating others. The authority of Kongzi the theoretician depended on Kongzi the master judge of character,
and the abstract statements ascribed to Kongzi in turn fortified the authority of specific judgments.149 Some Kongzi sayings even feature Kongzi abstracting theory from particular cases, as
in this excerpt from the conclusion to “Wu di de” (with parallels in the Han Feizi and Lunyu):
Kongzi said, “I once wished to evaluate people by their faces, but Mie Ming’s [=Ziyu 子羽]
faults changed that. Then I wished to evaluate people by their words, but Yu’s [=Zai Wo 宰
我] faults changed that. Then I wished to evaluate people by their demeanor, but Shi’s
[=Zizhang 子張] faults changed that.”
孔子曰：“吾欲以顏色取人，於滅明邪改之；吾欲以語言取人，於予邪改之；吾欲以容
貎取人，於師邪改之。”150
Sayings like this one highlight the connection between the discrete judgments preserved in texts
like the Zuozhuan and the theoretical statements of the Zhuangzi and Lunyu. The Kongzi of the
Da Dai Liji comes to reflect on the difficulties and methodologies of “evaluating others” (qu ren
取人) by observing the “speech” (yuyan 語言) and “demeanor” (rongmao 容貌) of specific individuals. The theoretical and the practical are then united in a psychologizing description of
Kongzi as having a persistent “desire” (yu 欲) to evaluate others.151
149.An example of Kongzi articulating criteria for identifying junzi 君子 (noble men), shengren 聖人 (sages), and
other ideal types is “Ai gong wen wu yi 哀公問五義” (“Duke Ai asked about the five principles”), book 40 of the
Da Dai liji (SBCK 1/5a–7a; KZJY 9.14/pp. 172–3).
150.Da Dai liji 62 (SBCK 7/4a; KZJY 6.9/pp. 17–19). For parallels, see also Han Feizi 50 (SBCK 19/9a; KZJY 9.45/
p. 189) and Lunyu 5/10.
151.Compare Lunyu 15/25: “The Master said, ‘With respect to other people, who have I criticized and who have I
praised? If I have praised another, I have examined him in some way. Such people are how the three dynasties [i.e.,
Xia, Shang, and Zhou] straightened and followed the Way” (子曰：吾之於人也，誰毀誰譽？如有所譽者，其有
所試矣。斯民也，三代之所以直道而行也。).
66
○ Comments on natural phenomena
A smaller number of Kongzi sayings apply Kongzi’s extraordinary powers of “hearing” (wen 聞)
and judgment to natural phenomena like droughts, strange creatures, and artifacts. Sun Xingyan
collected these sayings in the “Bo wu 博物” (“Knowledgeable about things”) chapter of the
Kongzi jiyu.
One of these comments is preserved in the “Bian wu 辨物” (“Discriminating things”) chapter
of the Shuiyuan:
Ji Huanzi had a well dug and came across an earthen jar, inside which was a sheep. He then
asked Kongzi about it, saying that he had come across a dog. Kongzi said, “According to what
I have heard, it is a sheep, not a dog. The oddities of wood and stone are the kui and
wangliang; the oddities of water are the long and the wangxiang; the oddity of earth is the fen
sheep.” Huanzi said, “Excellent!”
季桓子穿井得土缶，中有羊，以問孔子，言得狗。孔子曰：以吾所聞，非狗，乃羊也。
木石之怪夔罔兩，水之怪龍罔象，土之怪羵羊也，非狗也。桓子曰：善哉！152
The episode is unusual in that Ji Huanzi does not question Kongzi out of a sincere desire for advice or judgment. His sole intention is to test Kongzi’s knowledge, as if Kongzi’s reputation as
an expert on such matters was already established.
Another anecdote from the same Shuiyuan chapter features Kongzi extrapolating lessons
about understanding natural phenomena:
When King Zhao of Chu crossed the Yangzi there was an animal as big as a dipper that struck
the king’s boat head-on and stopped along the middle of the boat. King Zhao thought this
exceedingly strange and so sent a messenger to ask Kongzi about it. Kongzi said, “Its name is
‘duckweed fruit.’ You can split it open and eat it. Only a hegemon can find it—it is an
auspicious omen.” Afterwards there was a flying bird in Qi with a single leg that perched
before the palace and then stretched its wings and leapt up. The marquis of Qi thought this
exceedingly strange and so sent a messenger to ask Kongzi about it. Kongzi said, “Its name is
‘the Shang sheep’ and it is anxious to inform the people to quickly repair the irrigation ditches
because heaven will soon send a great rain.” Just as Kongzi had said, heaven then sent a great
152.Shuiyuan 18 (SBCK 18/17b; KZJY 11.2/p. 243). For parallels, see also Guoyu “Lu yu xia 魯語下” (SBCK 5/8b;
KZJY 11.1/p. 243), Shiji 47.1912, and Kongzi jiayu 16 (SBCK 4/11a).
67
rain that covered all the states in water, but Qi alone was safe. When Kongzi returned his
disciples asked him about it. Kongzi said, “They were extraordinary events! Children have a
nursery rhyme, ‘The king of Chu crosses the Yangzi and finds duckweed fruit, big as a palm,
red like the sun. Cut open and eaten it is as delicious as honey.’ This is [heaven’s] response to
Chu. Two children will also hold hands and face each other while hopping on one leg, saying,
‘When heaven is about to send a great rain, the Shang sheep will arise to dance.’ Now Qi has
found it—this, too, is [heaven’s] response. After a nursery rhyme there always follows a
response.” Thus a sage does not only guard the Way, he also looks into the accounts of natural
phenomena and soon finds their responses.
楚昭王渡江，有物大如斗，直觸王舟，止於舟中。昭王大怪之，使聘問孔子。孔子曰：
此名萍實，令剖而食之。惟霸者能獲之，此吉祥也。其後齊有飛鳥，一足，來下，止於
殿前，舒翅而跳。齊侯大怪之，又使聘問孔子。孔子曰：此名商羊，急告民，趣治溝
渠，天將大雨。於是如之，天果大雨。諸國皆水，齊獨以安。孔子歸，弟子請問。孔子
曰：異哉！小兒謡曰：楚王渡江得萍實，大如拳，赤如日，剖而食之美如蜜。此楚之應
也。兒又有兩兩相牽，屈一足而跳，曰：天將大雨，商羊起舞。今齊獲之，亦其應也。
夫謡之後，未嘗不有應隨者也。故聖人非獨守道而已也，睹物記也，即得其應矣。153
The Shuiyuan gives a decidedly ordinary explanation for Kongzi’s seemingly extraordinary
knowledge of natural phenomena. Just as the Kongzi of other Kongzi comments displays an
amazing knowledge of various received traditions, here he is revealed as a master of nursery
rhymes. Once again we find a Kongzi saying quoting a pre-existing tradition, extracting its wisdom, and applying that wisdom to new contexts. Kongzi’s knowledge of the natural world is
simply an extension of his general powers of wen 聞 (learning or hearing). As Duke Ai of Lu 魯
哀公 is made to say in a dialogue with Kongzi from the Xunzi, “If not you, Master, then no one
has heard of it” (非吾子無所聞之也。).154
● Kongzi commentaries
Kongzi commentaries are the other main sub-genre of dependent Kongzi sayings in my scheme.
153.Shuiyuan 18 (SBCK 18/17b; KZJY 11.12/p. 247)
154.Xunzi 31 (SBCK 28/21a; KZJY 6.23/p. 128). For a parallel discussion of Kongzi’s knowledge of natural
phenomena, see Nylan & Wilson 2010, pp. 89–91.
68
As noted earlier, the difference between a Kongzi comment and a Kongzi commentary lies in the
nature of the texts that are commented on. When explicating the Yi 易, Shi 詩, and Shu 書,
Kongzi sayings participated in textual genres that were far more prestigious than early anecdotal
literature, and probably also more stable. The status, circulation, and archaic language of the
classical traditions prompted Kongzi sayings which engaged their source texts in a much more
sustained and systematic fashion.
From a Han perspective, Kongzi’s role as commentator par excellence owed a great deal to
the view that Kongzi had a hand in the creation of the Five Classics. According to the “Kongzi
shijia” chapter of the Shiji, when Kongzi compiled the Shi he “discarded the duplicates and
adopted those which could applied to ritual and propriety” (去其重，取可施於禮義); with the
Shu and li 禮 (ritual texts) he “pursued the remnant rituals of the three dynasties [Xia, Shang, and
Zhou] and ordered the traditions of the Documents in chronicle form from the time of Yao and
Shun down to Duke Mu of Qin” (追跡三代之禮，序書傳，上紀唐虞之際，下至秦繆); with
the Music he “united the tones of the ‘Shao,’ ‘Martial Virtue,’ Elegentiae, and Eulogies” (合韶
武雅頌之音); and he “relied on scribal records to make the Chunqiu” (因史記作春秋).155 For
Han commentators, determining the true meaning of a classical text was a matter of understanding the lesson Kongzi had enshrined therein, as in a Kongzi jiayu episode prompted by an
episode from the Chunqiu in which the virtuous hegemon Duke Wen of Jin 晉文公 summons the
Zhou king to a meeting in an apparent breach of ritual propriety:
<Chunqiu>Heaven’s King [i.e., the King of Zhou] went on a hunt north of the He.
天王狩于河陽。156
155.Shiji 47.1935–6, 1943. On Kongzi’s association with the Chunqiu, see Kern 2011.
156.Xi gong 28 (SBCK 7/3a). This same line is also mentioned in the “Kongzi shijia” (Shiji 47.1943) as an example
of how Kongzi “restricted the phrasing [of the Chunqiu] to indicate widely applicable [lessons]” (約其文辭而指博).
69
<Zuozhuan>Zhongni said: As a subject to summon one’s lord—this cannot be a model lesson.
Thus was written, “Heaven’s King went on a hunt north of the He,” which still illuminates
[Duke Wen’s] virtue even though it was not the proper place [for a meeting between a duke
and a king].
仲尼曰：以臣召君[*kun]，不可以訓[*huns]。故書曰：天王狩于河陽。言非其地，也且
明德也。157
<Kongzi jiayu>Zigong asked Kongzi, “If Duke Wen of Jin really did summon the Son of
Heaven and made the vassal lords pay court at the same time, then why did you say that
‘Heaven’s King went on a hunt north of the He’ when you made the Chunqiu, Master?”
Kongzi said, “As a subject to summon one’s lord—this cannot be a model lesson. And [I]
merely wrote that [Duke Wen] led the vassal lords to serve the Son of Heaven.”
子貢問於孔子曰：晉文公實召天子，而使諸侯朝焉。夫子作春秋，云天王狩于河陽。何
也？孔子曰：以臣召君，不可以訓。亦書其率諸侯事天子而已。158
The premise of Zigong’s question in the Kongzi jiayu—“when you, Master, made the Chunqiu...”—is an explicit statement of the genre’s appeal, at least from a Han perspective: Kongzi
commentaries were a direct line to the ultimate source of the wisdom contained within the
classics.
However, it is also important to note that the perception of Kongzi as the author/compiler of
the Five Classics was virtually non-existent in pre-Han texts, the single glaring exception being
Mengzi 3B/9, which credits Kongzi with the creation of the Chunqiu.159 One striking passage
from the Shanghai Museum “Ji Kangzi wen yu Kongzi 季康子問於孔子” [“Ji Kangzi asked
Kongzi”] manuscript even has Kongzi quoting a certain Meng Ziyu’s 孟子餘 views on the value
of the Shi and Shu, as if Kongzi did not have the requisite authority to pronounce on canonical
texts:
Kongzi said, “I have heard that Meng Ziyu said, ‘The Shu make manifest the virtues of noble
men; the Shi commemorate the intentions of noble men; the Etiquette (?) constrains the
157.Zuozhuan Xi 28 (SBCK 7/8b; KZJY p. 413).
158.Kongzi jiayu 42 (SBCK 10/1a).
159.SBCK 6/13a.
70
conduct of noble men.’...”
孔子曰：丘聞之。孟子餘曰：夫書者，以著君子之德也；夫詩也者，以誌君子之志；夫
義(＝儀?)者，以謹君子之行也。160
Han-era authors invariably reserved such statements for Kongzi alone.
Rather than treat it as a single textual genre, it is perhaps more helpful to think of “commentary” as a collection of commentarial sub-genres distinguished by specific interpretive modes.
Early authors borrowed Kongzi’s voice for some of these modes more so than for others.
○ Amplifying commentaries
Kongzi’s versatility as a source of canonical commentary is readily apparent in the so-called
“Kongzi shilun 孔子詩論” (“Kongzi On the Shi”) manuscript from the Shanghai Museum corpus, a text whose Kongzi sayings exhibit multiple commentarial strategies. One of these strategies emerges from a comparison of the text’s multiple statements on the “Gan tang 甘棠”
(“Sweet Pear-Tree”) ode:
160.Shanghai bowuguan cang Zhanguo Chu zhushu vol 5, pp. 211–12.
71
“Kongzi shilun” strip #15:
“Kongzi shilun” strip #24:161
“Gan tang” longs for the man and respects
and cares for his tree.
Its protection is profound. The care of “Gan
tang” is due to the Duke of Shao.
[Kongzi said:...] In “Gan tang” I obtain respect
for the ancestral temple. People’s nature was
ever thus: when they greatly esteem the man they
invariably respect his position; when they delight
in the man they are invariably fond of his actions. When they hate the man it is also thus [i.e.,
they disrespect his position and dislike his
actions].”
[孔子曰：...]吾以（於）甘棠，得宗廟之敬。
甘棠〔思〕及其人，敬愛其樹。
民性固然：甚貴其人，必敬其位；
其保厚矣。甘棠之愛，以邵公〔之故也。 悅其人，必好其所為。惡其人者亦然。
Consider the two underlined eight-character lines from strips #15 and #24. Despite their differences, their parallel structure (X其人，敬其Y), parallel meaning (“long for” si 思 versus “esteem” gui 貴; “respect and care for” jing ai 敬愛 versus “respect” jing 敬), and joint association
with the “Gan tang” ode hint at some sort of genetic relationship. Yet there is a crucial difference
between them: the strip #24 version appears in a section prefaced by “Kongzi said” (Kongzi yue
孔子曰) but the strip #15 version does not. Moreover, parallels in a number of early texts generally confirm the attribution of the strip #24 line to Kongzi but not strip #15.162
Why was the strip #24 saying attributed to Kongzi but not the strip #15 saying? The answer
to that question leads us to a key feature of Kongzi yue commentaries and of Kongzi sayings generally. Not only is the strip #24 version significantly longer than the strip #15 version, it also
operates at a higher level of abstraction. The first such clue is the use of the word bi 必, whose
161.I follow Huang Huaixin’s (2004, pp. 18–22) reconstruction of the “Shilun” except where noted. Although I
prefer 保 to Huang’s 報 in this line, this is not an obvious choice. As I read it, the more obvious interpretation is that
the “protection” or “care” shown for the sweet pear-tree in the ode (“Do not lop it or knock it” 勿翦勿伐) mirrors
the “protection” the Duke of Shao showed to his people. On the other hand, that mirroring is also a good example of
“reciprocity” (bao 保).
162.For strip #15 parallels with no connection to Kongzi, see Zuozhuan Ding 9 (SBCK 28/4b) and Zuozhuan Xiang
14 (SBCK 28/4b), a Liu Xin 劉歆 memorial at Hanshu 73.3127, and Zheng Xuan’s 鄭玄 commentary on the “Gan
tang” ode (SBCK 1/15b).
72
insertion creates a general rule or maxim out of the more situation-specific version in strip #15:
“when [people] greatly esteem the man they invariably respect his position” (甚貴其人，必敬其
位) versus “‘Gan tang’ longs for the man and respects and cares for his tree” (〔思〕及其人，
敬愛其樹). The Kongzi yue version also replaces shu 樹 (“tree”) with wei 位 (“[official] position”), a metaphorical extension from “tree” to the notion of “where one plants oneself” or
“where one takes one stand,” i.e., one’s official position. Another clue is the phrase “people’s nature was ever thus” (min xing guran 民性固然), which inserts the word xing 性 (nature) to mark
the maxim as an abstract statement.163 The Kongzi saying then extracts the core structure of the
original version to generate a new parallel saying: “when [people] delight in the man they are invariably fond of his actions” (悅其人，必好其所為。). Finally, it extends the saying’s logic to
include the opposite case: “when [people] hate the man it is also thus [i.e., they do not respect his
position and they are not fond of his actions]” (惡其人者亦然。). The end result is a far more
gong 公 (impartial or abstract) and rhetorically sophisticated version of the eight-character line
from strip #15, which is then embedded in a longer paragraph with similarly structured and elaborated Kongzi comments.164
In sum, the “Kongzi shilun” author appears to have quoted Kongzi not for the content of the
saying but for its abstraction, amplification, and integration in a sequence of equally abstract Shi
commentary. The wisdom that is on display in the Kongzi saying is of a limited sort, a simple
163.This is section two in both Huang Huaixin’s and Li Xueqin’s reconstructions. For a translation of the latter, see
Jiang Guanghui 2008; for a comparison of various proposed reconstructions, see Xing Wen 2008.
164.Not coincidentally, Kongzi’s comments on three other “Bangfeng” 邦風 odes in the same section of the
“Shilun” — “Ge tan” 葛覃 (“The Cloth-Plant Spreads”), “Mu gua” 木瓜 (“A Quince”), and “Di du” 杕杜 (“Tall
Pear-Tree”) — follow the same pattern, with a similarly structured saying following the phrase min xing guran 民性
固然: “when [people] see something beautiful they invariably wish to trace it back to its root” (見其美，必欲反其
本) for “Ge tan”; “hidden thoughts will invariably find some means of expression” (其隱志，必有以（俞）抒也)
for “Mu gua”; “when [people] cleave to what they care for they invariably say, ‘How can I abandon this?’” (離其所
愛，必曰吾奚捨之) for “Di du.”
73
generalization from the concrete situation of strip #15 to the abstract (or gong 公) lesson of strip
#24. From a pedagogical perspective, it would seem that this type of Kongzi commentary teaches
its audience not what to think about the Shi, but how to mine the Shi for generalizable lessons
and human universals.
○ “Greater meaning” (da yi 大義) commentaries
The “Lun shu 論書” (“Assessing the Documents”) chapter of the Kong congzi suggests a useful
label for another sub-genre of commentary in a passage featuring Zixia 子夏 asking Kongzi to
explain the da yi 大義 (“greater meaning”) of the Shu:
Zixia asked about the greater meaning of the Shu. The Master said, “In the Canons of the
Emperors I see the sageliness of Yao and Shun; in the Counsels of the Great Yu, the Counsels
of Gao Yao, and the Yi Ji I see the loyalty, diligence, achievement, and merit of Yu, Ji, and
Gao Yao; in the Luo Proclamation I see the virtue of the Duke of Zhou. Thus, in the Canons
of the Emperors one can observe excellence; in the Counsels of the Great Yu and the Tribute
of Yu one can observe the handling of affairs; in the Counsels of Gao Yao and Yi Ji one can
observe good governance; in the Great Plan one can observe due order; in the Great
Declaration one can observe what is appropriate; in the Five Proclamations one can observe
humaneness; and in the Lu Punishments one can observe prohibitions. When these seven
things are comprehended, the greater meaning of the Shu is upheld.
子夏問書大義。子曰：吾於帝典見堯舜之聖焉；於大禹、皋陶謨、益稷見禹、稷、皋陶
之忠勤功勳焉；於洛誥見周公之德焉。故帝典可以觀美，大禹謨、禹貢可以觀事，皋陶
謨、益稷可以觀政，洪範可以觀度，泰誓可以觀義，五誥可以觀仁，甫刑可以觀誡。通
斯七者，則書之大義舉矣。165
Unlike, say, the (pseudo-)Kong Anguo 孔安國 Shangshu commentary or the typical modern
commentary, which assists one’s reading of the text by glossing problematic words and phrases,
this Kongzi commentary shows little interest in substantively engaging the content of the Shu.
Instead, the “greater meanings” supplied by Kongzi establish interpretive guidelines for each Shu
165.Kong congzi 2 (SBCK 1/6a). The continuation of this particular passage includes more da yi commentary.
74
text, the core ideas which come to frame the interpretation of individual lines, phrases, and
words. Although not labeled as such, the most well known example of a “greater meaning” commentary from the early period is perhaps the Mao Shi 毛詩 (Mao Odes). When the Mao preface
to “Guan ju 關雎” (“Fish-hawks”) states that the ode is about “the virtue of the consort” (houfei
zhi de ye 后妃之德也),166 it is no coincidence that the Mao commentary subsequently interprets
individual words and phrases in keeping with the global evaluation, e.g., by understanding the binome yaotiao 窈窕 as a description of a virtuous woman.167 Statements about a text’s “greater
meaning” do not derive from early interpreters’ close readings of texts; they render interpretation
a fait accompli.168
The appeal of “greater meaning” Kongzi commentaries in the Han likely owed something to
the perception that the proliferation of commentarial traditions represented a corruption of
canonical wisdom. Ban Gu 班固 (32–92 CE), an author whom the Hou Hanshu 後漢書 (History
of the Later Han) praised as “having simply explained the greater meaning without engaging in
chapter-and-verse commentary” (不為章句，舉大義而已。),” explicitly connects Kongzi with
the notion of “greater meaning” in the preface to the “Yiwenzhi 藝文志” bibliography:
Formerly, after Zhongni passed away his subtle words were cut off, and after the seventy
disciples died the greater meaning was distorted. Thus the Chunqiu was split into five
[traditions], the Shi into four [traditions], and the Yi into traditions of several experts. Amid the
strategic scheming of the Warring States the true and the counterfeit came into conflict and the
words of the various masters became chaotic and confused.
昔仲尼沒而微言絕，七十子喪而大義乖。故春秋分為五，詩分為四，易有數家之傳。戰
166.SBCK 1/1b.
167.Kern 2007, pp. 783–87.
168.For other examples, see the Zhouyi “Xici zhuan” (SBCK 8/6b–7b; KZJY p. 450) and Huainanzi 18 (SBCK
18/2b; KZJY 4.34/p. 60). The nearly word-for-word parallels between Kongzi’s da yi commentary on the Shi in the
Hanshi waizhuan 2 (SBCK 2/17b; KZJY 4.43/pp. 63–64) and his Shu commentary in the Shangshu dazhuan (SBCK
5/13a–b; KZJY 4.44/pp. 64) and Kong congzi 2 (SBCK 2/7a) even suggests that it mattered little what corpus was
being commented on so long as its meaning was said to be profound.
75
國從衡，真偽分爭，諸子之言紛然殽亂。169
The text goes on to lament the further fracturing of the “greater meaning” during the Warring
States and Qin periods before praising Han emperors’ efforts to reunify the tradition. Writing towards the end of the Eastern Han, Xu Gan 徐幹 (170–217) developed these ideas further when
he argued for the superiority of da yi learning over a pedantic interest in terminology and word
glosses:
In learning, the greater meaning (da yi 大義) comes first and terminology comes after.
Terminology follows after the elucidation of the greater meaning. Nonetheless, shabby Ru in
their broad learning labor over terminology. They are particular about vessels and implements
and they are boastful about their word glosses. They extract their chapter-and-verse
commentaries without being able to synthesize the implications of the greater meaning and
thereby grasp the heart of the former kings. This is no different from a woman or a scribe
reciting the Shi, or a houseboy passing along orders. And so this leads scholars to exert their
intellects on empty matters and to be ignorant of the Way, and to spend day and night in
fruitless pursuits. Thus the noble man always chooses his teacher accordingly.
凡學者，大義為先，物名為後。大義舉而物名從之。然鄙儒之博學也，務於物名，詳於
器械，矜於詁訓，摘其章句，而不能統其大義之所極以獲先王之心。此無異乎女史誦
詩，內豎傳令也。故使學者勞思慮而不知道，費日月而無成功。故君子必擇師焉。170
From this perspective, Kongzi’s da yi commentary acquires a greater intellectual historical significance. It is a genre which cuts through the accumulated clutter of recent textual traditions to
articulate the original, core meaning of the classics. In contrast to the pedantic prolixity of Han
“chapter-and-verse” (zhang ju 章句) commentaries, the pithiness and generality of “greater
meaning” commentaries were markers of the ancient, the authentic, and the profound, all the
more so when attributed to Kongzi.171
Although describing what is likely a late fourth-century BCE text in this way is probably
169.Hanshu 30.1701.
170.Zhonglun 1, “Zhi xue 治學” (SBCK 4b–5a).
171.Yan Kejun’s 嚴可均 (1762–1843) statement in the preface to the Kongzi jiyu that non-canonical Kongzi sayings
also preserve Kongzi’s “greater meaning” testifies to the longevity of this idea in the Chinese tradition. See Kongzi
jiyu jiaobu pp. 1–2.
76
anachronistic, the “Kongzi shilun” contains several examples of what might be labeled “greater
meaning” commentary, including the following:
Kongzi said, “In ‘Ge tan’ I grasp thoughts of the clan’s origins (?)...
In ‘Gan tang’ I grasp respect for the ancestral temple...
In ‘Mu gua’ I grasp the indispensability of money and silk...
In ‘Di Du’ I grasp submission to rank (?)...
孔子曰：吾以（於）葛覃，得氏初之詩（志）。...
吾以（於）甘棠，得宗廟之敬。...
吾以（於）木瓜，得幣帛之不可去也。...
吾以（於）杕杜，得雀（爵）〔服之〕。...172
A similar formula appears elsewhere in “Kongzi shilun” in a passage which expresses Kongzi’s
personal responses to a series of Shi:
Kongzi said, “’Wan qiu’—I praise it; ‘Yi jie’—I delight in it; ‘Shi jiu’—I trust it; ‘Wen
wang’—I find it excellent; ‘Qing miao’—I respect it; ‘Lie wen’—I am pleased by it...”
孔子曰：宛丘吾善之，猗嗟吾喜之，鳲鳩吾信之，文王吾美之，清廟吾敬之，烈文吾悅
之...173
As Chen Tongsheng and Huang Huaixin have noted, the use of the formula “in [text X] I see Y”
(wu yu X jian Y 吾於X見Y) in the “Kongzi shilun” and other texts hints at a tradition of Kongzi
yue commentary from the Warring States through the Han period.174 Direct engagement with the
substance of these Shi texts seems to be beside the point. In the “Gan tang” ode, for instance,
there is no mention of an “ancestral temple” (zong miao 宗廟) or any hint that Duke Shao was
venerated as an ancestor. These are the commentarial equivalents of the global character assessments observed in Kongzi comments on people, statements which fix the interpretation of texts
to an overarching value or concept.175 Moreover, the association of that concept with a particular
172.Huang Huaixin 2004, p. 19.
173.Huang Huaixin 2004, p. 21.
174.Huang Huaixin 2004, pp. 282–315; Chen Tongsheng 2004, pp. 220–21.
175.See Kern 2010, pp. 37–38, for the suggestion that the broad categories attached to specific Shi in the “Kongzi
shilun” might explain “the confusingly wide range of situations to which the ‘Airs’ are applied in early
historiography.”
77
ode might have had less to do with the content of the ode itself than with the need to map a matrix or list of concepts onto the Shi as a whole.
The “Kongzi shilun” examples also share with the Kong congzi “greater meaning” commentary an insistent use of the first-person pronoun—wu 吾. But we need not conclude from this that
wu 吾 was a mark of authenticity or a record of Kongzi’s actual experience of the Shi.176 By having Kongzi speak in the first person, the commentary establishes a more intimate connection between Kongzi and the Shi, thereby modeling the kind of affective relationship students were expected to have with these texts. To truly understand the Shi, the “Kongzi shilun” author seems to
suggest, a student must experience for himself a properly regulated “joy” or “faithfulness” or the
like.177 By shifting the focus from the texts to the learner, these “greater meaning” commentaries
revealed the study of canonical texts as an occasion for self-cultivation.
○ “Clarifying” (ming 明) commentaries
The zi yue sayings preserved in the “Wen yan 文言” (“Patterned words”) commentary to the first
two hexagrams of the received Zhouyi feature a more familiar commentarial strategy, at least
from the perspective of the later tradition:
The first yang says: “A submerged dragon does not act.” What does this mean? The Master
says: “This refers to one who has a dragon’s virtue yet remains hidden. He neither changes to
176.Foucault’s (1979, p. 152) comments on the diverse functions of the first-person pronoun are instructive in this
context: “The self that speaks in a treatise on mathematics—and that indicates the circumstances of the treatise’s
composition—is identical neither in its position not its functioning to the self that speaks in the course of a
demonstration, and that appears in the form of ‘I conclude’ or ‘I suppose.’ In the first case, the ‘I’ refers to an
individual without an equivalent who, in a determined place and time, completed a certain task; in the second, the ‘I’
indicates an instance and a level of demonstration which any individual could perform provided that he accept the
same system of symbols, play of axioms, and set of previous demonstrations.” The function of Kongzi’s “I” of the
“Kongzi shilun” is close to this latter sense.
177.For a parallel reading of the hermeneutics revealed in early manuscripts, see Kern 2003 & 2007.
78
suit the world nor seeks fulfillment in fame. He hides from the world but does not regret it,
and though this fails to win approval, he is not sad. When he takes delight in the world he is
active in it, and when he finds it distresses him he turns his back on it. One who is resolute in
his unwillingness to be uprooted is a submerged dragon.”
初九曰：潛龍勿用。何謂也？子曰：龍德而隱者也。不易乎世，不成乎名，遯世無悶，
不見是而無悶，樂則行之，憂則違之，確乎其不可拔，潛龍也。178
In his third-century CE commentary to the “Wen yan,” Wang Bi 王弼 (226–249) characterized
this passage as follows: “This entire section uses the affairs of men to clarify [the text]” (此一章
全以人事明之也。). Ming 明 (“to clarify”) is an apt label for what distinguishes this Kongzi
commentary from those discussed above because it is the first Kongzi commentary considered in
this section which introduces a concrete concept—the yinzhe 隱者 (one who is hidden)—to anchor an enigmatic image—the qian long 潛龍 (submerged dragon])—to a specific context—the
realm of human affairs. Here the choice of commentary probably reflects the nature of source
material. The Ci Fei anecdote from the Lüshi chunqiu did not require a Kongzi comment in order
to be understood because its content was more or less transparent. In contrast, the enigmatic line
statements of the Zhouyi require some frame of reference before they could be made sense of and
applied to real-world contexts.179
Another example of a clarifying commentary is Kongzi’s discussion of the notion of “min zhi
fumu 民之父母” (father and mother to the people) from the “Jiong zhuo 泂酌” (“At the Wayside
Pool”) ode, versions of which appear in the Shanghai Museum “Min zhi fumu” manuscript, the
“Kongzi xianju 孔子閒居” chapter of the Liji, and the “Lun li 論禮” chapter of the Kongzi jiayu.
178.Zhouyi (SBCK 1/4a; KZJY p. 440). Translation adapted from Lynn 1994, p. 132. Chen Tongsheng (2004, p. 60)
also compares the “Kongzi shilun” Kongzi comments with the “Wen yan” to suggest that Kongzi yue statements
represent “the principal form of an early stage of pre-Qin Ru canonical commentary” (先秦儒家早期解經的主要形
式).
179.Another example is Mawangdui “Er san zi wen 二三子問” (“The disciples asked”) text, in which Kongzi
elucidates the recurring dragon image in the Zhouyi. See Csikszentmihalyi 2002, pp. 137–138, for a discussion of
this passage. For a transcription, see Chen Songzhang & Liao Mingchun 1993.
79
Below are the first few lines of the Liji version:
Kongzi was resting in leisure with Zixia in attendance when Zixia said, “May I ask about the
Shi [#251] that says ‘All happiness to our lord / Father and mother to the people?’ What sort of
person can be called ‘father and mother to the people?’”
Kongzi said, “The ‘father and mother of his people’ must have insight into the source of
ritual and music, and must achieve the ‘five reaches’ and practice the ‘three withouts.’ Thus
[his virtue] will pervade the world. When there is some loss within the four directions he will
always know it in advance. Such a man is referred to as ‘father and mother to the people.’”
孔子閒居，子夏侍。子夏曰：敢問詩云：凱弟君子，民之父母。何如斯可謂民之父母
矣？孔子曰：夫民之父母乎，必達於禮樂之原，以致五至而行三無，以橫於天下。四方
有敗，必先知之。此之謂民之父母矣。180
Here again, the Kongzi saying cherry-picks a single line from a canonical text and gives it a specific frame of reference. I shall have more to say about this example below (pp. 108–111).
The “Xici zhuan” commentary to the received Zhouyi furnishes another zi yue comment that,
at first glance, resembles a clarifying commentary:
Tongren [Fellowship, Hexagram 13] says: “First howling and wailing, but afterwards there is
laughter.” The Master said: “In the Way of the noble man,
There’s a time for going forth / And a time for staying still,
A time to remain silent / And a time to speak out.
When two people share their hearts, / Their sharpness severs metal,
And the words of those who share their hearts, / Are fragrant like orchids.
同人：先號咷而後笑。子曰：君子之道。
或出或處(*k-hlaʔ) / 或默或語(*ŋaʔ)。
二人同心(*səәm) / 其利斷金(*kəәm)。
同心之言(*ŋan) / 其臭如蘭(g-rân)。181
There is sometimes a fine line between introducing a new concept in order to clarify a source
text, and using a source text as a prompt to introduce a tangential idea. “Wen yan” and “Kongzi
xian ju” walk this line more successfully than this “Xici zhuan” zi yue saying, whose only
connection to the line statement in question seems to be the word tong 同 (same). This also happens to be the most ostentatiously rhymed zi yue saying in the entire Zhouyi and perhaps even in
180.Liji 29 (SBCK 15/9b–11a; KZJY pp. 545–6). See also Shanghai bowuguan cang Zhanguo chu zhushu vol. 2, pp.
154ff and Kongzi jiayu 27 (SBCK 6/17b–18b).
181.Zhouyi “Xici zhuan 繫辭傳” (SBCK 7/6a; KZJY p. 445). Translation from Lynn 1994, p. 58.
80
the entire corpus of Kongzi sayings. One wonders if the compilers of the “Xici zhuan” cared less
about explaining the canon than using it as a prompt for other texts they were interested in.
○ Word and character glosses
A series of Kongzi yue sayings preserved in the Shuowen jiezi 說文解字 (Explanations of Simple
and Complex Characters) feature Kongzi engaged in yet another commentarial sub-genre, the
explication of individual words and characters:182
[The entry for wang 王:] Kongzi said, “‘One’ piercing ‘three’ makes ‘king.’”
孔子曰：一貫三為王。183
[The entry for quan 犬:] Kongzi said, “Looking at the character for quan, it resembles a
drawing of a dog.”
孔子曰：視犬之字，如畫狗也。184
[The entry for he 貉:] Kongzi said, “The character he is pronounced like ‘bad’ (e 惡).”
孔子曰：貉之為言惡也。185
Without additional context it is impossible to know whether these Kongzi sayings had a context
which has been lost to us. One possibility is that they were extracted from longer anecdotes, as in
this dialogue from the “Ji yi 祭義” (“The Meaning of Sacrifice”) chapter of the Liji:
When Zhongni performed the autumn sacrifice he presented the offerings and advanced, in his
person he was sincere and his movement was brisk with many steps. When the sacrifice was
finished, Zigan [=Zigong] asked, “When you speak about the sacrifice, Master, [you say that
one should be] jiji and qieqie. But in the sacrifice now you were neither jiji nor qieqie. Why is
that?” The Master said, “Jiji has to do with demeanor; [it means] distant. Qieqie [also] has to
do with demeanor; [it means] turning inward. When one’s demeanor is distant it is as though
one turns back into oneself. But how can one successfully relate to the spirits [in this way],
and how can one be distant and turned inward [when conducting sacrifices]? After gifts are
182.Michael Nylan (Nylan & Wilson 2010, p. 88) mentions such glosses and notes “the popularity of punning
glosses in Han times, visual and verbal, which are thought to exhibit the tight correspondences threading through all
phenomenal existence.”
183.KZJY 5.94/p. 93.
184.KZJY 5.102/p. 95.
185.KZJY 5.102/p. 95.
81
returned and the music has concluded the offering vessels are brought forth, the ritual and
music is arranged, and there is a full complement of the hundred officials. If a junzi becomes
jiji and qieqie then how can he lose himself in the sacrifice? How can my words have only one
facet? Everything has that to which it is suited.
仲尼嘗：奉薦而進，其親也愨，其行也趨趨以數，已祭。子贛問曰：子之言祭，濟濟漆
漆然。今子之祭，無濟濟漆漆。何也？子曰：濟濟者，容也，遠也。漆漆者，容也，自
反也。容以遠，若容以自反也。夫何神明之及交？夫何濟濟漆漆之有乎？反饋樂成，薦
其薦俎，序其禮樂，備其百官，君子致其濟濟漆漆。夫何慌惚之有乎？夫言豈一端而
已？夫各有所當也？186
Given the scarcity of standalone Kongzi yue glosses outside the Shuowen jiezi, it is difficult to
imagine these glosses circulating independently of episodes like this one.
○ Meta-commentaries
In my discussion of Kongzi comments on people I noted that a number of Kongzi sayings dealt
with the practice, the problems, and the possibilities of character evaluation in a more general
sense. In a few Kongzi commentaries one also encounters sayings which address the practice of
commentary itself, as in the following examples from the “Xici zhuan” and “Kongzi shilun”:
<“Xici zhuan”>The Master said, “Writing does not completely express one’s words, and
words do not completely express one’s meaning. This being the case, is the meaning of the
sages opaque?” The Master said, “The sages established the figures [i.e. the hexagrams] to
completely express their meaning, they set forth the hexagrams to completely express what is
and what can be made to be, and they appended phrases to completely express their words”
子曰：書不盡言，言不盡意，然則聖人之意，其不可見乎？子曰：聖人立象以盡意，設
卦以盡情偽，繫辭以盡其言。187
<“Kongzi shilun”>Kongzi said, “Shi have no hidden intentions; Music188 has no hidden
emotions; patterned expression has no hidden words.”
186.Liji 23 (SBCK 14/6a–b; KZJY pp. 538–39).
187.Zhouyi “Xici zhuan” (SBCK 7/10b; KZJY p. 447). The unnecessary repetition of zi yue in this passage is curious
and may indicate that these markers were added after the the composition of the underlying treatise.
188.Here I follow Scott Cook’s suggestion (1995, p. 21–22) in his translation of the “Yue ji” 樂記 (“Record of
Music”) to capitalize “Music” (yue 樂) in contrast with less morally edifying “music” (yin 音).
82
孔子曰：詩無隱志，樂無隱情，文無隱言。189
Neither of these sayings articulates a fully developed hermeneutics or theory of commentary. But
attached to other Kongzi yue and zi yue commentaries, the confident assertion of the readability
of the Zhouyi and “Kongzi shilun” enhances the authority of the commentator and the wisdom of
his commentary. If canonical texts “completely express [the sages’] meaning” (jin yi 盡意) and
“have nothing to hide” (wu yin 無隱), then students can hope to unlock the wisdom of these texts
with the proper guidance.190 The pervasiveness of Kongzi yue/zi yue comments and commentary
made Kongzi an altogether appropriate cheerleader for the commentarial endeavor and traditional learning generally.191 Such statements may have also been an antidote to the Zhuangzian criticism that received textual traditions were merely “the dregs of the ancients” (古人之糟魄), faint
shadows of the sages’ original ideas.192
Independent Kongzi sayings
In most of the examples considered thus far, a Kongzi yue or zi yue saying is contingent on the
text it amplifies, epitomizes, or explicates. As we have seen, “dependent” sayings are often rather
uninformative insofar as they guide the audience’s response to a source text without introducing
189.Whether or not the strip numbered #1 by Ma Chengyuan 馬承源 (Shanghai bowuguan cang Zhaoguo Chu
zhushu vol. 1, p. 123) truly belongs at the head of the “Kongzi shilun,” its placement there seems reasonable given
that the idea expressed in this saying lays the theoretical groundwork for Shi interpretation just as the zi yue
comments in the “Xici zhuan” licenses that text’s interpretation of the Yi hexagrams.
190.However, the reconstruction of the character I read as yin 隱 is problematic. Here again I follow Huang Huaixin
(2004, pp. 267–68).
191.Although not a commentary per se, see also p. 90 below for Kongzi’s assertion in Zhuangzi book 22 that “the
ancient resembles the present” (gu you jin 古猶今), a statement which underwrites the study of ancient models.
192.Zhuangzi 13 (SBCK 5/34a).
83
new ideas of their own. In contrast, “independent” Kongzi sayings do not respond to other texts,
and the lack of pretext lends independent sayings an air of authority, as if they were worthy of
being recorded and remembered simply because Kongzi said them. Kongzi yue or zi yue is the
only context such sayings require.
Far and away the most famous independent Kongzi sayings are those found in the Lunyu.
Consider Lunyu 1/1:
The Master said, “To regularly practice what one has learned—is this not a pleasure? To have
friends come from afar—is this not a joy? To not resent going unrecognized by others—is this
not noble?”
子曰：學而時習之，不亦說乎？有朋自遠方來，不亦樂乎？人不知而不慍，不亦君子
乎？
As is typical, the Lunyu makes no effort to explain what might have prompted this Kongzi saying
or why it was deemed important enough to be the first entry in the entire collection. For the Lunyu compilers and their audience alike, Kongzi did not require an introduction more elaborate
than “The Master said,” presumably of their overriding interest in Kongzi himself. By my count,
219 of the 513 (43%) Lunyu entries consist of independent sayings like Lunyu 1/1. Another 110
entries (21%) feature Kongzi responding to questions from disciples or other interlocutors. However, forty-three of these entries follow a simple format in which the interlocutor asks a simple
indirect question (e.g., Lunyu 2/6: “Meng Wubo asked about filial piety” 孟武伯問孝。) as a
prompt for Kongzi’s views on various topics. While such mini-dialogues are vaguely reminiscent
of dependent Kongzi comments, in fact they seem to serve a very different purpose. Less important than the “independent” and “dependent” labels themselves is how we answer the question,
“What is a given saying for?” If the answer is “to present Kongzi’s ideas” and not “to assist
one’s understanding of another text,” then the saying is best understood as an “independent”
saying.
84
The following example from Lüshi chunqiu 26/2, “Wu da 務大” (“Striving after the great”),
is the longest independent Kongzi saying in any ostensibly pre-Han text:
Confucius said, “Swallows compete for good nesting sites beneath the eaves. Mother birds
feed their fledglings, which coo happily to one another and feel themselves secure. If there is a
crack in the chimney, the fire will spread along the rafters. Yet the swallows will not change
their manner—why is that? Because they are unaware that disaster is about to overtake them.
Is this not stupid? Rare is the ministerial officer who avoids being such a birdbrain. Most
officials increase their titles, emoluments, fortunes, and honors. Fathers and sons, older and
younger brothers, form cliques within a single state, cooing happily to one another, and
thereby endanger their altars to the soil and grain. They are unaware of how close they are to a
crack in the chimney, and so in the final analysis their wisdom is no greater than that of
swallows. Therefore it is said: When the whole world is in a state of total anarchy, no country
is safe; when a whole state is in utter chaos, no family is secure; and when the whole family is
in total disorder, no individual is secure; and when the whole family is in total disorder, no
individual is secure. This expresses my meaning. Therefore, the safety of the small inevitably
depends on that of the large, and the safety of the large inevitably depends on that of the small.
Only when the small and the large, noble and base, cooperate by assisting each other will all
find happiness.”
孔子曰：燕爵爭善處於一屋之下，母子相哺也，區區焉相樂也，自以為安矣。灶突決，
上棟焚，燕爵顏色不變，是何也？不知禍之將及之也，不亦愚乎！為人臣而免於燕爵之
智者寡矣。夫為人臣者，進其爵祿富貴，父子兄弟相與比周於一國，區區焉相樂也，而
以危其社稷，其為灶突近矣，而終不知也，其與燕爵之智不異。故曰：天下大亂，無有
安國；一國盡亂，無有安家；一家盡亂，無有安身，此之謂也。故細之安，必待大；大
之安，必待小。細大賤貴，交相為贊，然後皆得其所樂。193
This saying is very different from the dependent sayings considered above, both with respect to
its length and its rhetorical strategy. None of our earlier examples argue as this one does, articulating a position ex nihilo by first introducing it with a colorful analogy and then reinforcing it
with an apt quotation. And unlike the zi yue sayings of the Lunyu, whose striking lack of argumentation fuels the impression that Kongzi’s ideas are authoritative simply by virtue of being
Kongzi’s, this saying seems to derive its authority not from Kongzi yue but from the argument
itself.
As it happens, the same saying also has a word-for-word parallel in section 13/7 of the Lüshi
chunqiu, “Yu da 諭大” (“Illustrating the great”). However, there it is attributed not to Kongzi but
193.Lüshi chunqiu 26/2 (SBCK 26/3b–4a; KZJY 7.21/p. 153). Translation from Knoblock & Riegel 2000, p. 648.
85
to a certain Jizi 季子: “Jizi said, ‘Swallows compete for good nesting sites beneath the eaves...’”
(季子曰：燕雀爭善處於一屋之下...).194 Chen Qiyou 陳奇猷 has argued from a comparison
with Jizi material in the Han Feizi that the quotation was mistakenly attributed to Kongzi by
some later Lüshi chunqiu editor.195 The merits of Chen Qiyou’s argument aside, the relative
scarcity of argumentative Kongzi sayings in pre-Han texts is further reason not to trust its attribution to Kongzi.
The presentation of Kongzi sayings in some texts likewise indicates that their authors were
sensitive to the quotability of their texts. Recall that Youzi’s critique in the “Tan gong” (p. 46)
lent some support ot my distinction between dependent and independent Kongzi sayings. Also
consider the Shanghai Museum “Lu bang da han 魯邦大旱” (“There was a great drought in Lu”)
manuscript (Appendix 1:C), which frames its Kongzi sayings as responses to a series of questions posed by two interlocutors: Duke Ai of Lu 魯哀公, who seeks Kongzi’s advice on ending a
drought, and the disciple Zigong 子貢, who asks Kongzi to justify that advice. One can easily
imagine some of these sayings circulating independently, perhaps as statements on the importance of rectifying punishments and virtue or on the ignorance of the masses. But “Lu bang da
han” solves the problem of what to quote by having Kongzi offer his own moral in the conclusion: “Ah, Ci [=Zigong]! I will tell you: a junzi considers supplication to be a matter of patterning, whereas the masses consider it a matter of the spirits” (賜也，我告爾：命也者，君子
以為文，庶民以為神。). Probably not coincidentally, the very same maxim happens to appear
in book 17 of the Xunzi, “Tian lun 天論” (“Discussing heaven”), as an anonymous saying.196 As
194.Lüshi chunqiu 13/7, “Yu da 諭大” (“Illustrating the Great”) (SBCK 13/13b–14a).
195.Lüshi chunqiu xin jiaoshi 13.733–34, n. 22.
196.Xunzi book 17 (SBCK 11/22b).
86
with the Kongzi comments discussed above in which Kongzi enjoins his disciples to ji 記 (take
note of) or memorialize some lesson (p. 51), here the Kongzi saying reveals the “Lu bang da
han” author’s in supplying his audience with an quotable, independent dictum
The contrast between the Kongzi sayings of “Lu bang da han” and Zigong’s critique of those
sayings is also instructive. Zigong, a disciple with a reputation for verbal artistry,197 presents an
argument in support of the view that sacrifices are entirely unnecessary, and thus unwarranted.
Ironically enough given the respect accorded Kongzi in the episode, Zigong’s perfectly balanced
speech at 98 characters is longer than all of Kongzi’s statements combined. By comparison, the
Kongzi sayings themselves are short, pithy, and less overtly crafted. Their relative simplicity is
perhaps one indication that the “Lu bang da han” author thought of Kongzi as an authority whose
ideas did not require ornamentation. True authorities delivered pronouncements; mere persuaders
crafted arguments.
● Objects of commentary
Kongzi sayings as objects of commentary stand in stark contrast to Kongzi comments and commentary. When used as an object of commentary, it is the Kongzi saying which creates the occasion for explication or amplification, and this function determines the form and even the content
of the saying. Generally speaking, Kongzi comments and commentary had to be readily comprehensible in order to be effective, otherwise they contributed nothing to the understanding of the
197.See, e.g., Lunyu 11/3 and also the definitive statement in the “Zhongni dizi liezhuan” (Shiji 67.2195) that
“Zigong was a silver-tongued and clever speaker 子貢利口巧辭.” An extended anecdote from the same chapter
(Shiji 67.2197–2201) also provides a clear demonstration of Zigong’s powers of persuasion.
87
text being commented on. In contrast, Kongzi sayings as objects of commentary have to challenge the audience’s comprehension in some way in order to justify the commentary, as when the
terms jiji and qieqie, two examples of the notoriously impenetrable class of vocabulary of reduplicating binomes, prompted Kongzi’s glosses in his dialogue with Zigong in the Liji (p. 81).198
Of course, many Kongzi sayings eventually received commentaries by virtue of having been included in prestigious texts like the Lunyu or Liji. But certain sayings anticipated their own explication more than others.
The impenetrability of some Kongzi sayings did not go unnoticed in the early period. In the
discussion of Lunyu 14/25 in his treatise “Wen Kong 問孔” (“Interrogating Kongzi”), Wang
Chong wondered aloud about Kongzi’s failure to properly explain himself:
<Lunyu 14/25>Qu Boyu sent a messenger to Kongzi. Kongzi sat with the messenger and said,
“How does your master fare?” The messenger responded, “My master desires to minimize his
errors but is unable to do so.” After the messenger had departed, Kongzi said, “That
messenger! That messenger!”
蘧伯玉使人於孔子。孔子與之坐而問焉，曰：夫子何為？對曰：夫子欲寡其過而未能
也。 使者出。子曰：使乎！使乎！
<Lunheng>There was a point to Kongzi’s criticism. But without clarifying the error he merely
said, “That messenger! That messenger!” Later generations suspected that Kongzi was
confused because they did not understand how the messenger had erred. Han Feizi said,
“When the writing is spare the followers spar.” Why was Kongzi so sparing when he said,
“That messenger!”?
所非猶有一實，不明其過，而徒云使乎使乎！後世疑惑，不知使者所以為過。韓子曰：
書約則弟子辨。孔子之言使乎，何其約也？199
Wang Chong’s quotation of the Han Feizi is fitting in this context. There the relevant passage
reads:
When writing is spare the followers spar; when the laws are abridged the people litigate. This
is why a sage’s writings are always clearly expounded and an enlightened ruler’s laws are
always meticulously detailed.
198.A definitive statement on the reduplicating binome is is Knechtges 1987, pp. 1–13. Martin Kern (2007, p. 784)
has described these terms as “more volatile than any other kind of word in early texts.”
199.Lunheng 28 (SBCK 9/15a).
88
書約而弟子辯，法省而民訟簡。是以聖人之書必著論，明主之法必詳事。200
Wang Chong went on to argue that Kongzi’s response—“That messenger! That messenger!”—
was actually a criticism of the messenger’s willingness to reveal his master’s failings. But Wang
Chong’s real target in this passage was Kongzi, who deserved some blame for not having stated
his critique more clearly. For those familiar with the context of Wang Chong’s Han Feizi quotation, the implication must have been obvious: Kongzi was not speaking like a true sage when he
uttered these words.
For an author like Wang Chong who prized clear and simple prose, inscrutable language and
the verbal wrangling it encouraged were things to be avoided. But many other early authors seem
to have realized the potential of “sparing” language to engage audiences and consciously cultivated a laconic style. Some, like the author of the following Huainanzi episode (with parallels in
the Lüshi chunqiu, Wenzi, Shuiyuan, and Liezi), even celebrated such language as an ideal form
of communication known as weiyan 微言 (subtle speech):
Duke Bai asked Kongzi, “Is it possible for men to use subtle speech?” Kongzi did not respond.
Duke Bai said, “What about throwing a rock into water?” Kongzi said, “The divers of Wu and
Yue could get it.” Duke Bai said, “What about throwing water into water?” Kongzi said, “At
the confluence of the Zi and Sheng rivers, Yi Ya could taste the water and tell them apart.”
Duke Bai said, “If so, then is it not the case that men inherently cannot share subtle speech?”
Kongzi said, “Why say they cannot? Only those who understand the meaning of words [can
share subtle speech]. Those who understand the meaning of speech do not speak with words.
Those who wrangle fish get wet, and those who chase after animals must run—it is not
because they enjoy it. Thus the highest speech eschews speaking, and the highest action is
non-action. Those who shallowly understand what it is they’re wrangling for are worthless.”
Duke Bai did not grasp this, and so he died at Yushi.
白公問於孔子曰：人可以微言？孔子不應。白公曰：若以石投水中，何如？曰：吳越之
善沒者能取之矣。曰：若以水投水，何如？孔子曰： 澠之水合，易牙嘗而知之。白公
曰：然則人固不可與微言乎？孔子曰：何謂不可？誰知言之謂者乎！夫知言之謂者，不
以言言也。爭魚者濡，逐獸者趍，非樂之者也。故至言去言，至為無為。夫淺知之所爭
者末矣。白公不得也，故死於浴室。201
200.Han Feizi 47 (SBCK 18/7a).
201.Huainanzi 12 (SBCK 12/1b–2a; KZJY 16.24/pp. 371–2). For the Lüshi chunqiu parallel, see Lüshi chunqiu 18/3
89
The anecdote is both an endorsement and an illustration of weiyan. After Kongzi declines to respond to Duke Bai’s straightforward question, the duke must resort to metaphor to elicit
Kongzi’s thoughts on the subject, to which Kongzi responds with some metaphors of his own.
(In the Lüshu chunqiu version of the anecdote, the final Kongzi saying ends with the line “Only
those who understand the meaning of words [can share subtle speech]” and so does not attribute
the conclusion to Kongzi. In the later Huainanzi, the concluding comment of the Lüshi chunqiu
compilers migrated into the Kongzi saying itself.) The moral of the anecdote, that “the highest
speech eschews words” (至言去言), brings to mind Kongzi’s and Laozi’s advice from the “man
of Chu lost his bow” anecdote (p. 57) to “get rid of” (qu 去) specific words in the original sayings. This was a fitting ideal for authors who sought to harness the rhetorical potential of enigmatic language.202
Also note that the failure to “grasp” (de 得) Kongzi’s words reflects poorly on the interlocutor. In a Kongzi anecdote from book 22 of the Zhuangzi, weiyan similarly functions as a test of
the hearer’s “spirit” (shen 神):
Ran Qiu asked Zhongni, “Can [the time] before heaven and earth be known?” Zhongni said,
“It can. The ancient resembles the present.” Ran Qiu dropped his questioning and retired but
the next day he again came to see [Kongzi], saying, “Master, yesterday when I asked whether
[the time before] heaven and earth could be known you said, ‘It can. The ancient resembles the
present.’ Yesterday I understood it clearly but now it is unclear. May I ask what you meant?”
Zhongni said, “Yesterday it was clear because your spirit had already accepted it. Today it is
unclear because you seek it without using your spirit....”
冉求問於仲尼曰：未有天地可知邪？仲尼曰：可。古猶今也。冉求失問而退。明日復
見，曰：昔者吾問未有天地可知乎？夫子曰：可。古猶今也。昔日吾昭然，今日吾昧
然。敢問何謂也？仲尼曰：昔之昭然也，神者先受之；今之昧然也，且又為不神者求
(SBCK 18/7a–b; KZJY 16.23/p. 371). I read wei 唯 for shei 誰 in the line 誰)知言之謂者乎 in keeping with the
Lüshi chunqiu version.
202.Compare also Lunyu 7/8 with its claim that Kongzi “would not repeat himself if, after raising one corner, [his
student] could not respond with the other three” (舉一隅不以三隅反，則弗復也。). See also Laozi 70 (SBCK 下/
16b–17a): “When those who understand me are few then I am honored. Thus a sage dresses shabbily and hugs his
jade tight” (知我者希，則我者貴，是以聖人被褐懷玉。).
90
邪！203
In the continuation of the dialogue Kongzi goes on to explain the meaning of the three-word
statement that “the ancient resembles the present” (gu you jin 古猶今). More important for our
purposes than the substance of Kongzi’s explanation is the fact that Kongzi’s meaning is opaque
without it. Once again we find non-comprehension presented as the mark of an uncultivated
mind, a notion which places the onus on the audience to make sense of opaque weiyan.
Three additional examples of Kongzi sayings as objects of commentary can be found at Lunyu 2/5, 4/15, and 15/3:
<Lunyu 2/5>When Meng Yizi asked about filial piety, the Master said, “Nothing disobedient.”
The Master reported this exchange to his driver, Fan Chi, saying, “When Mengsun asked me
about filial piety, I responded, ‘Nothing disobedient.’” Fan Chi said, “What did you mean?”
The Master said, “While alive, serve [one’s parents] with ritual; after they have died, bury
them with ritual and make offerings to them with ritual.”
孟懿子問孝。子曰：無違。 樊遲御，子告之曰：孟孫問孝於我，我對曰：無
違。樊遲曰：何謂也？子曰：生，事之以禮；死，葬之以禮，祭之以禮。
<Lunyu 4/15>The Master said, “Zengzi! My Way has a single thread running through it.”
Zengzi said, “Indeed.” After the Master had left his followers asked Zengzi, “What did he
mean?” Zengzi said, “The Master’s way is nothing but loyalty and empathy.”
子曰：參乎！吾道一以貫之。曾子曰：唯。子出，門人問曰：何謂也？曾子曰：夫子之
道，忠恕而已矣。
<Lunyu 15/3>The Master said, “Ci, do you suppose that I know what I do because I have
studied many kinds of things?” He replied, “I do. Is that not the case?” “It is not. I bind it all
together with a single [thread].”
子曰：賜也，女以予為多學而識之者與？對曰：然，非與？曰：非也，予一以貫之。
In our discussion of Kongzi sayings as commentary we saw that the question he wei ye 何謂也
(What does it mean?) often prefaced Kongzi comments on canonical texts. Here the same question explicitly marks Kongzi’s words—“nothing disobedient” and “My Way has a single thread
running through it”—as statements requiring explication. But in fact the question is entirely
203.Zhuangzi 22 (SBCK 7/54a–b; KZJY 17.18/p. 394).
91
redundant. Particularly in Lunyu 4/15 and 15/3, the mere mention of a single idea that unlocks
the secret of Kongzi’s Way raises the question: what is it? Kongzi’s unnamed disciples in Lunyu
4/15 voice the only possible response to Kongzi’s cryptic utterance.204 Wang Chong seems to
have realized this when he observed of Lunyu 2/5, “If Fan Chi had not asked for an explanation
of ‘nothing disobedient,’ then he would not have understood” (使樊遲不問，毋違之說，遂不
可知也。).205 We can restate Wang Chong’s point even more strongly: the Kongzi sayings are
essentially, and probably intentionally, incomprehensible without additional clarification.
The following passage from the Han Feizi (with parallels in the Kongzi jiayu, Shuiyuan, and
Shangshu dazhuan) features Kongzi commenting on his own sayings in order to defend himself
against the charge of inconsistency:
When Zigao, the Duke of She, asked Zhongni about governance, Zhongni said, “Governance
depends on pleasing those nearby and attracting the distant.” When Duke Ai asked Zhongni
about governance, Zhongni said, “Governance depends on selecting accomplished men.”
When Duke Jing of Qi asked Zhongni about governance, Zhongni said, “Governance depends
on moderating expenses.” When the three dukes had departed, Zigong asked, “The three dukes
all asked the same question about governance, so why did you give different answers?”
Zhongni said, “The capital of Ye is large but the state is small and its people have rebellion in
their hearts. That is why I said that governance depends on pleasing those nearby and
attracting the distant. Duke Ai of Lu has three great ministers. Outside his borders he is
obstructed by the vassal lords and neighboring armies, and within he is beset by those who
204.Contrast this approach with that of Bryan Van Norden (2002, pp. 222–3), who puzzles over the “anomalies” of
Lunyu 4/15: “To begin with, Confucius announces summarily that his Way has ‘one thread,’ and when Zengzi
concurs, Confucius walks out of the room. Immediately, disciples rush up to ask Zengzi for an interpretation. I find
this somewhat off in itself for a simple reason: Zengzi was not an especially bright guy. Indeed, he is described in
11:18 as ‘stupid’ (lu 魯). In general, if we look at passages other than 4:15, Zengzi seems to have a strong personal
commitment to the Way, but he never comes across as someone who was particularly acute intellectually. There is
never, for example, a passage in which Zengzi is commended by Confucius for his insight, in the manner that
Confucius does commend Zixia 子夏 (3:8) or Zigong (1:15). Why, then, does Confucius throw something patently
cryptic at Zengzi, and then leave the room? And why do the other disciples assume that Zengzi will know what the
Master meant? Finally, note that Confucius says his Way has ‘one thread.’ However, when asked what Confucius
meant, Zengzi mentions two things. Interpreters tend to assume that the ‘one thread’ has two aspects, but it is odd
that Zengzi never made clear what the ‘one thread’ was.” Van Norden eventually concludes that Zengzi’s
explanation was a later interpolation, a conclusion which finds some support in my discussion of Lunyu Zengzi
material in chapter three (pp. 261–265). But Van Norden’s “anomalies” vanish when one ceases to treat Lunyu 4/15
as a realistic depiction of an actual historical event and instead consider the rhetorical imperatives at work in this and
similar scenes.
205.Lunheng 28 (SBCK 9/2b).
92
would make a fool of their lord. It is surely these three ministers who will ensure that the
ancestral temple is not maintained and that the state altars go without sacrifices. That is why I
said that governance depends on selecting accomplished men. Duke Jing of Qi built the Yong
Gate and the Luqin Tower, and in one morning he gave land worth three-hundred chariots to
three people. That is why I said that governance depends on moderating expenses.”
葉公子高問政於仲尼，仲尼曰：政在悅近而來遠。哀公問政於仲尼，仲尼曰：政在選
賢。齊景公問政於仲尼，仲尼曰：政在節財。三公出，子貢問曰：三公問夫子政一也，
夫子對之不同，何也？仲尼曰：葉都大而國小，民有背心，故曰政在悅近而來遠。魯哀
公有大臣三人，外障距諸侯四鄰之士，內比周而以愚其君，使宗廟不掃除，社稷不血食
者，必是三臣也，故曰政在選賢。齊景公築雍門，為路寢，一朝而以三百乘之家賜者
三，故曰政在節財。206
Kongzi’s response to Zigong’s quite reasonable question echoes Youzi’s distinction in “Tan
gong” between what the Master said and what he said it about (p. 46). As in “Tan gong,” the key
to resolving the apparent contradiction lies in understanding that these seemingly independent
sayings are actually dependent comments on the failings of the three dukes. In effect, Kongzi argues that his sayings should not be interpreted as independent, generally applicable statements on
governance.
In his discussion of this and similar examples, Oliver Weingarten has argued that the
episode’s author “sought to resolve the contradiction that arises from the co-presence of the various statements by integrating an explanation into the narrative of the text in the form of an utterance that he laid into the Master’s mouth.”207 This is a plausible account of the motivation behind
the anecdote. Parallels of the saying “governance lies in moderating expenses” appear in a number of anecdotes involving Kongzi and Duke Jing of Qi, including in Lunyu 13/16, which might
suggest that these sayings did indeed circulate independently in the early period before being
206.Han Feizi 38 (SBCK 16/2b; KZJY 10.41/p. 226–27). For a parallel, see Shuiyuan 7 (SBCK 7/9b). Weingarten
(2010, pp. 116ff) discusses this example in his analysis of Lunyu 11/22, which also features Kongzi resolving
apparent contradictions between sayings.
207.Weingarten 2010, p. 130.
93
brought together in the Han Feizi anecdote.208 On the other hand, the phrases “pleasing those
nearby and attracting the distant” (悅近而來遠 yue jin er lai yuan), “selecting accomplished
men” (xuan xian 選賢), and “moderating expenses” (jie cai 節財) are so widespread in the early
corpus that they are perhaps best read as stock concepts of early intellectual discourse (Appendix
2:U). Rather than read the episode as an effort to resolve contradictions among pre-existing
Kongzi sayings, the utterly generic nature of the sayings in question might suggest that they were
attributed to Kongzi by the author of this anecdote precisely in order to create the perception of
contradiction. Like Lunyu 4/15, the episode might have been designed in order to prompt
comment.
● Objects of criticism
Kongzi sayings as objects of criticism are functionally similar to Kongzi sayings as objects of
commentary insofar as authors who criticized Kongzi also tended to treat his words as independent sayings, which they then used to prompt discussions of various kinds. However, these
Kongzi sayings differ from the types discussed above insofar as they challenge the image of
Kongzi as a figure of supreme wisdom and virtue.209
The Kongzi saying disputed by Youzi in “Tan gong” (p. 46) may be one instance of a Kongzi
saying designed to be criticized: “losing one’s position one should wish for swift poverty; losing
one’s life one should wish for swift decay” (喪欲速貧，死欲速朽). As Ziyou 子游 explains, the
offending Kongzi saying derives from two dependent Kongzi comments:
208.See Shiji 47.1911 and Kongzi jiayu 13 (SBCK 3/16a).
209.Regier (2010, p. 39) labeled this phenomenon “weaponized” quotation.
94
When he dies,
Once out of office,
the quicker he decays the better.
the quicker he becomes poor the better.
死不如速朽之愈也。
喪不如速貧之愈也。
⤷
⤶
Losing one’s position one should wish for swift poverty;
losing one’s life one should wish for swift decay.
喪欲速貧，死欲速朽。
Ziyou’s argument is bolstered by the observation that the two Kongzi comments appear independently in the Kongzi jiayu,210 which might indicate that the resulting Kongzi saying was created
by mashing the Kongzi comments together into a single parallel couplet. But as Youzi’s critique
indicates, the end result is simply too bizarre to be taken seriously. Zheng Xuan 鄭玄 (127–200)
put this point succinctly in his commentary to “Tan gong”: “poverty and decay are not what people desire” (貧、朽，非人所欲。).211 Given Youzi’s and Zheng Xuan’s responses, my suspicion is that the saying in question was designed precisely in order to elicit a negative reaction.
The Han Feizi passage above in which Kongzi explains his seemingly contradictory responses to the three lords is a more obvious example of a Kongzi saying created as an object of criticism. Although the episode itself is not critical of Kongzi, the Han Feizi follows it with harsh
critique of Kongzi’s advice: “There are those who say, ‘Zhongni’s responses were state-ruining
sayings’” (或曰：仲尼之對，亡國之言也。).212 The Han Feizi author goes on to suggest his
own “single saying which could have removed the three dukes’ troubles: ‘know your underlings’” (一言而三公可以無患，知下之謂也).213 The episode’s polemical context might suggest
that what ultimately motivated the anecdote was a desire not to defend Kongzi, but to make him
look foolish.
210.Kongzi jiayu 42 (SBCK 10/1a–2a).
211.Liji 3 (SBCK 2/15a).
212.Han Feizi 38 (SBCK 16/2b).
213.Han Feizi 38 (SBCK 16/4a).
95
The Mozi 墨子 is another text that does not shy away from criticizing Kongzi and his teachings, as seen in the following episode:
Zigao the Duke of She asked Zhongni about governance, saying: “What sort of person is good
at governing?” Zhongni responded, “Someone who is good at governing attracts the distant
and renews the old.” Mozi learned of this and said, “Zigao did not understand what to ask, and
Zhongni did not understand how to answer. How could Zigao not know that those who are
good at governing attract the distant and renew the old? He should have asked how this is
accomplished. [Kongzi] didn’t tell [Zigao] what he didn’t know, he told him what he knew.
That is why Zigao did not understand what to ask, and Kongzi did not understand how to
answer.”
葉公子高問政於仲尼曰：“善為政者若之何？”仲尼對曰：“善為政者，遠者近之，而舊
者新之。”子墨子聞之曰：“葉公子高未得其問也，仲尼亦未得其所以對也。葉公子高豈
不知善為政者之遠者近也，而舊者新是哉？問所以為之若之何也。不以人之所不智告
人，以所智告之，故葉公子高未得其問也，仲尼亦未得其所以對也。”214
Here Mozi fills the role usually assigned to Kongzi, the authority who somehow “learns of” (wen
聞) and subsequently comments on happenings which he could not have witnessed firsthand.215
But the Kongzi saying in question is too generic to be taken seriously. Sayings built on the yuan
遠/jin 近 (near/far) polarity pervade early texts, including in the “Shang xian 尚賢” (“Exalting
accomplished men”) chapter of the Mozi: “those nearby, keep them content; those far away,
make them come home to you” (近者安之，遠者歸之).216 The prevalence of that idea in the early literature leads me to suspect that the author of this passage attributed an extremely conventional saying to Kongzi precisely in order to accuse him of “not understanding how to answer” (未
得其所以對).
The “Fei Ru 非儒” (“Against the Ru”) chapter of the Mozi attributes another contrived saying to Kongzi as part of an anecdote about his travails between Chen and Cai:
When Kong Qiu was in dire straits between Cai and Chen, he had only simple broth without
any grain. After ten days, Zilu boiled a pig for him, and Kong ate it without asking where the
214.Mozi 46 (SBCK 11/14b–15a; KZJY 13.21/p. 281).
215.This use of Mozi is rather rare, even within the Mozi. For the two other instances of “Mozi learned of this and
said” in that text, see Mozi book 49 (SBCK 13/4b & 13/9a).
216.Mozi 10 (SBCK 2/18a).
96
meat came from. Zilu stole another’s clothing and sold it to buy wine, and Kong Qiu drank it
without asking where it came from. But when Duke Ai received Kongzi, he refused to sit on a
mat that was not properly arranged and refused to eat meat that was not correctly cut. Zilu
came forward to ask, “Why do you act differently now compared to when you were between
Chen and Cai?” Kong Qiu said, “Come, I shall tell you: at that time we were worried about
our survival, and now we are worried about what is proper.”
While starving and destitute he did not refuse what was wrongly acquired in order to stay
alive, and when full and well off he faked his conduct in order to make a show of himself.
What could be more depraved and hypocritical than this?!
孔丘窮於蔡陳之間，藜羹不糂，十日，子路為享豚，孔丘不問肉之所由來而食；號人衣
以酤酒，孔丘不問酒之所由來而飲。哀公迎孔丘，席不端弗坐，割不正弗食，子路進，
請曰：何其與陳、蔡反也？孔丘曰：來！吾語女：曩與女為苟生，今與女為苟義。夫飢
約則不辭妄取，以活身，贏飽則偽行以自飾，汙邪詐偽，孰大於此！217
One need not be told that Kongzi’s response to Zilu appears nowhere else in the extant literature
to see that the author of this passage likely fabricated a blatantly hypocritical Kongzi saying for
polemical purposes.
Other authors were more subtle about their irreverence for Kongzi, opting for parody over
outright polemic. One such example is this exchange between Yan Hui and Zhongni from chapter six of the Zhuangzi 莊子, “Da zongshi 大宗師” (“The Great Ancestral Teacher”):
Yan Hui said, “I have progressed.” Zhongni said, “What do you mean?” Yan Hui said, “I have
forgotten humaneness and propriety.” Zhongni said, “Fine, but you are not there yet.” Another
day Yan Hui again came to see Zhongni and said, “I have progressed.” Zhongni said, “What
do you mean?” Yan Hui said, “I have forgotten ritual and music.” Zhongni said, “Fine, but
you are still not there.” Yet another day Yan Hui came to see Zhongni and said, “I have
progressed.” Zhongni said, “What do you mean?” Yan Hui said, “I sit and forget.” Zhongni
looked troubled and said, “What do you mean by ‘sitting and forgetting’?” Yan Hui said, “I let
my limbs and body fall away and my awareness slip away. Leaving my body and losing my
wits, I share in the great interpenetration [of things]. This is what I mean by ‘sitting and
forgetting.’” Zhongni said, “Sharing in [the great interpenetration of things] you have no
preferences, and being in flux you have no constancy. Thus you have become accomplished.
Please allow me to become your follower.”
顏回曰：回益矣！仲尼曰：何謂也？曰：回忘仁義矣。曰：可矣，猶未也。它日復見
曰：回益矣。曰：何謂也？曰：回忘禮樂矣。曰：可矣，猶未也。它日復見曰：回益
矣。曰：何謂也？曰：回坐忘矣。仲尼蹴然曰：何謂坐忘？顏回曰：墮枝體，黜聰明，
離形去知，同於大通，此謂坐忘。仲尼曰：同則無好也，化則無常也，而果其賢乎？丘
也請從而後也。218
217.Mozi 9 (SBCK 9/22a).
218. Zhuangzi 6 (SBCK 3/26a–b; KZJY 16.31/p. 378). For a parallel, see Huainanzi 12 (SBCK 12/14a; KZJY 16.31/
97
For anyone familiar with the depiction of Kongzi in the Lunyu, a text for which ren 仁 (humaneness), yi 義 (propriety), li 禮 (ritual), and yue 樂 (music) are core concepts, Kongzi’s advice to
“forget” (wang 忘) these values is the first clue that the anecdote might fall outside mainstream
Kongzi quotation practice. Another is Kongzi’s repetition of the question “What do you mean?”
(he wei ye 何謂也). As noted in the discussion of Kongzi sayings as objects of commentary,
Kongzi’s interlocutors typically posed this question to Kongzi, not the other way around. But
here it is Yan Hui’s cryptic utterances, especially zuo wang 坐忘 (“sitting and forgetting”), that
must to be explained to Kongzi. The final clue is the anecdote’s ironic conclusion, which inverts
the Kongzi/Yan Hui relationship and turns master into disciple and disciple into master.219
At the same time, the Zhuangzi anecdote also shares certain features with mainstream Kongzi
sayings and anecdotes. At no point does the anecdote question Kongzi’s status as an exemplary
judge of character. Three times Yan Hui reports his progress to Kongzi, and three times Kongzi
offers an accurate assessment. Even in the conclusion, Kongzi displays an extraordinary humility
and talent for character evaluation in the course of recognizing and submitting to Yan Hui’s superior wisdom. The not-entirely-unsympathetic portrayal separates this parody from the crude
caricatures of the Mozi.
● Proof texts
A “proof text” is a quotation designed to enhance the persuasiveness or authority of the quoting
p. 378).
219.An anecdote from Zhuangzi 28 (SBCK 9/26a; KZJY 1.22/p. 8) ends with a similar result. Jean Levi (1995, pp.
41–65) in an article on the “subversive” use of citation in classical Chinese literature points to another example from
Zhuangzi book 14 (SBCK 5/49a–51a).
98
text. Proof texts qualify as independent sayings insofar as their content is not contingent on some
other text; in order to be useful as a proof text, a saying had to be detachable from its earlier context. Consequently, quoting authors typically ignored Youzi’s advice and treated Kongzi sayings
as maxims of general applicability.
The only Kongzi proof text in the entire Zuozhuan appears in the entry for the sixteenth year
of Duke Ai’s 哀公 reign (r. 494–477 BCE), the year of Kongzi’s death:
Summer. The fourth month. Jichou. Kong Qiu died. The Duke eulogized him as follows:
“Merciful heaven does not come to my aid, nor does it leave me even one old man to support
me, The One Man, in my position. Grievously I mourn. Woe is me! Alas! Father Ni, there is
no one to guide me.” Zigan [=Zigong] said, “Duke Ai shall not meet his end in Lu! The
Master had a saying, ‘Erring in ritual matters breeds ignorance, and erring in titles breeds
transgression.’ Erring in your aims is [also] ignorant, and error is the root of transgression.
While [The Master] was alive [my lord] could not employ him, but in death he eulogizes him.
This is not ritually proper. He refers to himself as ‘The One Man,’ but this is not his proper
title. My lord is doubly in error.”
夏，四月，己丑。孔丘卒。公誄之曰：旻天不弔，不憖遺一老，俾屏余一人以在位，煢
煢余在疚。嗚呼！哀哉！尼父無自律。子贛曰：君其不沒於魯乎。夫子之言曰：禮失則
昏，名失則愆。失志為昏，失所為愆。生不能用，死而誄之，非禮也。稱一人，非名
也。君兩失之。220
As we have seen, the received Zuozhuan often appeals to Kongzi as a source of comments on
anecdotes and people. But Zigong’s speech here is the first and only instance in the text in which
a character invokes Kongzi as an authority in his own right. Not coincidentally, it is also the only
Kongzi saying in the Zuozhuan which is not directed at a specific narrative. Zigong’s speech
marks the passing of Kongzi as a master judge and virtuoso of applied wisdom, and the emergence of Kongzi as a venerable source of quotable wisdom. On the one hand, Kongzi’s death
heralds the end of the more dynamic practice of commenting on contemporary affairs; on the
other, it crystallizes his sayings as authoritative teachings of general import.221
220.Zuozhuan Ai 16 (SBCK 30/7a; KZJY p. 426).
221.For a parallel discussion of this episode, see Schaberg 2001, pp. 307–10.
99
The most spectacular use of Kongzi yue proof texts in the early corpus appears in a set of
memorials ostensibly submitted to Han Wudi 漢武帝 in ca. 134 BCE by Dong Zhongshu 董仲舒
(179–104 BCE).222 The sixteen Kongzi sayings woven into these texts create the impression that
Dong Zhongshu’s advice meshed seamlessly with Kongzi’s teachings, as in the following excerpt which features a word-for-word quotation of Lunyu 13/12 :
Your subject has heard that when Yao received the mandate, the entire world was his concern
and he never took delight in his position. Thus he executed and expelled disorderly subjects
and he strived to find accomplished men and sages. This is how he obtained Shun, Yu, Ji, Xie,
and Gao Yao. All sages supplemented his virtue and accomplished and capable men assisted
him in his duties. His enlightening influence marched greatly onward, and the world was
harmonious. The myriad peoples were at peace in humaneness and delighted in propriety. All
obtained their proper places, their every action was in keeping with ritual, and they kept to the
Way with ease. Thus Kongzi said, “If there is a true king, it will take a generation before [the
people] become humane.” This is what [I] mean.
臣聞堯受命，以天下為憂，而未以位為樂也，故誅逐亂臣，務求賢聖，是以得舜、禹、
稷、卨、咎繇。眾聖輔德，賢能佐職，教化大行，天下和洽，萬民皆安仁樂誼，各得其
宜，動作應禮，從容中道。故孔子曰如有王者，必世而後仁，此之謂也。223
“This is what [I] mean” is my translation of ci zhi wei ye 此之謂也, a standard proof text marker
which highlights the convergence between the quoting and quoted texts, a connection reinforced
by the insertion of gu 故 (thus) immediately before Kongzi yue. Dong Zhongshu’s agreement
with Kongzi is not coincidental or superficial; at least rhetorically, the implicit claim is that he
has tapped into the thinking which motivated Kongzi to utter this saying in the first place.
Imperial edicts were the most prominent texts to use Kongzi sayings as proof texts. The first
edict below dates to 7 BCE during the reign of Emperor Ai 哀帝 (r. 7–1 BCE), and the second to
26 CE during the reign of Emperor Guangwu 光武帝 (r. 25–57 CE):
<Emperor Ai>It is when worldly customs become decadent and extravagant that the melodies
of Zheng and Wei arise. When decadent, those below are disobedient and the state is
impoverished. When baroque, too many people chase after trivialities and neglect what is
essential. When the melodies of Zheng and Wei arise, licentious and wicked influences spread.
222.For a discussion of the authenticity of these documents, see pp. 177–181.
223.Hanshu 56.2508.
100
In these circumstances desiring the people to be simple and frugal is akin to demanding a pure
stream when the source has been muddied. How could that not be difficult? Did not Kongzi
say, “Do away with the melodies of Zheng. The melodies of Zheng are licentious.” Let the
Music Bureau be abolished.
惟世俗奢泰文巧，而鄭衛之聲興。夫奢泰則下不孫而國貧，文巧則趨末背本者眾；鄭衛
之聲興則淫辟之化流，而欲黎庶敦樸家給，猶濁其源而求其淸流，豈不難哉？孔子不云
乎：放鄭聲，鄭聲淫。其罷樂府官。224
<Emperor Guangwu>When We think of the cruelty inflicted by oppressive officials—their
extreme use of punishments and their imprisonment of many wrongly accused people—We
are greatly grieved. Did not Kongzi say, “When punishments are wrongly applied the people
have nowhere to place their hands and feet”? Let those with a salary of at least 2,000 shi, the
various grandees, and the court deliberators discuss a reduction of punishments.
惟酷吏殘賊，用刑深刻，獄多寃人，朕甚愍之。孔子不云乎：刑罰不中，則民無所措手
足。其與中二千石、諸大夫、博士、議郎議省刑法。225
It is no accident that the Kongzi sayings quoted in these edicts also appear in the Lunyu. Proof
texts presumably had to have achieved a certain level of recognizability and authority before they
could function as proof texts. As we will see in the next chapter, tracking the use of Kongzi sayings as proof texts in edicts and similar contexts is one way to gauge the relative authority of the
sources of Kongzi sayings.
● Records of Kongzi’s life and thought
Kongzi sayings as records differ from the Kongzi sayings discussed above insofar as a text’s perceived value as a historical source depends as much, if not more, on its reception as on any intrinsic features.226 A Kongzi saying as comment or proof text is a comment or proof text by
224.Hanshu 22.1072–3.
225.Hou Hanshu 1.29.
226.Owen 2007, p. 7: “If we think of ‘authorship’ as a property of a text, like a title, then we can see that in many
cases it was something added by inference, just as titles so often were.”
101
virtue of its function in the text in which it appears. But a Kongzi saying could become a historical “record” simply if later readers chose to treat it as such. An example of this phenomenon is
book ten of the Lunyu, which has been read as a “record of Kongzi’s words and deeds while in
the villages of Lu” (記孔子在魯國鄉黨中言行).227 Were it not for the fact that (a) this material
appears in the Lunyu, and (b) “Kong” and “zi” are the first two characters of Lunyu 10/1, one
would struggle to discern any connection whatsoever to Kongzi, let alone evidence that book ten
was a historical record of some sort.228
Nevertheless, some sayings raise the issue of historicity more directly. After the Lunyu, the
one text most often appealed to as a source for the historical Kongzi is Shiji book 47, the
“Kongzi shijia 孔子世家.” The secondary status of the “Kongzi shijia” in Kongzi studies is ironic given that, of these two texts, only the “Kongzi shijia” actually presents itself as a biography
which ties Kongzi’s doings and utterances to a definite and realistic timeline. The concluding
comment to the “Zhongni dizi liezhuan” biography of Kongzi’s disciples also casts Sima Qian as
a critical historian who assessed various ideas about Kongzi against the available sources in order to “get close to the truth” (jin shi 近是).229 Earlier I observed that the biography reflected
Sima Qian’s interest in depicting Kongzi “as a man” (qi wei ren 其為人) and thus anchored
Kongzi sayings to specific moments in Kongzi’s life (p. 28). That intent is also evident in the
contrast between, e.g., Lunyu 7/23 and its “Kongzi shijia” parallel:
<Lunyu 7/23>The Master said, “Heaven birthed the virtue in me, so what is Huan Tui to me?”
子曰：天生德於予，桓魋其如予何？
227.This is the Xing Bing’s 邢昺 (931–1010) sub-commentary on the title of Lunyu book 10. See Lunyu zhengyi
10/1a (p. 86).
228.Weingarten 2009, p. 5.
229.For this quotation, see p. 184.
102
<“Kongzi shijia”>When Kongzi left Cao and went to Song, he practiced rituals under a great
tree with his disciples. Sima Huan Tui of Song wished to kill Kongzi, and so he pulled up
[Kongzi’s] tree. Kongzi’s disciples said, “We should [leave] quickly.” Kongzi said, “Heaven
birthed the virtue in me, so what is Huan Tui to me?”
孔子去曹適宋，與弟子習禮大樹下。宋司馬桓魋欲殺孔子，拔其樹。孔子去。弟子曰：
可以速矣。孔子曰：天生德於予，桓魋其如予何！230
Although the Shiji version of the Kongzi saying parallels Lunyu 7/23 word-for-word, it also contextualizes the saying in a way that makes the Lunyu version seem an altogether inadequate
record: nowhere does the Lunyu explain Huan Tui’s identity or his desire to kill Kongzi. It is
possible that the lack of context reflects the story’s popularity at the time of the Lunyu’s compilation, in which case the Lunyu compilers did not see a need to explain its details. On the other
hand, while a number of early texts mention the incident of “the tree being chopped down in
Song” (fa shu yu Song 伐樹於宋) in connection with Kongzi, the Shiji is the earliest extant text I
am aware of to cast Huan Tui as a principal character in the story. In any event, the comparison
with the Shiji reveals the Lunyu compilers’ relative disinterest in matters of historical context, as
if providing a historical record was not their primary motivation.
Lunyu 7/1, 15/6, and 17/19 raise the issue of historicity in a very different way:
<Lunyu 7/1>Insofar as I transmit without initiating and trust in and love the ancient, I would
humbly compare myself to Old Peng.
子曰：述而不作，信而好古，竊比於我老彭。
<Lunyu 15/6>Zizhang asked about following [the Way]. The Master said, “If your words are
loyal and trustworthy and your actions sincere and respectful, then you will follow it even in
foreign lands. But if your words are not loyal and trustworthy and your actions are not sincere
and respectful, then will you follow it even in your local area? While standing you must see it
facing you, while riding you must see it on the handle-bar. Only then will you follow [the
Way].” Zizhang wrote this down on his sash.
子張問行。子曰：言忠信，行篤敬，雖蠻貊之邦，行矣。言不忠信，行不篤敬，雖州
里，行乎哉？立則見其參於前也，在輿則見其倚於衡也，夫然後行。子張書諸紳。
230.Shiji 47.1921. Other references to this event state that Huan Tui “chopped down” (fa 伐) the tree, which seems
likelier than the Shiji version. See, e.g., Lüshi chunqiu 14/6 (SBCK 14/17b; KZJY 13.28/p. 287).
103
<Lunyu 17/19>The Master said, “I want to do without words.” Zigong said, “But if you do not
speak, Master, then what will your followers transmit?” The Master said, “What does Heaven
ever say? The four seasons march onward and the hundred things continue to be born. What
does Heaven ever say?”
子曰：予欲無言。子貢曰：子如不言，則小子何述焉？子曰：天何言哉？四時行焉，百
物生焉，天何言哉？
All of these passages have been cited as evidence in support of the Han bibliographers’ view that
the Lunyu was compiled by closest Kongzi’s disciples from their own records of Kongzi’s teachings. Lunyu 7/1 and 17/19 have been read as expressions of an ideological commitment to “transmit” (shu 述) the Master’s teachings verbatim,231 and Lunyu 15/6 as an example of a saying actually recorded by one of Kongzi’s followers.232 However, references to disciples writing down and
transmitting Kongzi’s words are by no means unique to the Lunyu. As noted elsewhere (p. 51 &
Appendix 1:B), a number of extant anecdotes in a range of early texts feature Kongzi urging his
followers to “take note of” (ji 記/zhi 誌) his words. It seems just as reasonable to treat such passages as generic features of the representation of master-disciple relations in the early literature,
not as evidence of reliable textual transmission.
“Imputed words” (Yu yan 寓言)
As noted earlier, Sun Xingyan in the final section of the Kongzi jiyu included a number of
Kongzi anecdotes under the label “imputed words” (yu yan 寓言), the implication being that
their authors were unconcerned with questions of historical accuracy and simply used Kongzi as
a vehicle for their own ideas. In so doing, Sun Xingyan did not just undermine the value of these
231.Liu Xinfang 2003, p. 4.
232.Chen Tongsheng 2004, p. 51.
104
particular sayings and anecdotes; he also indirectly guaranteed the historical value of the rest of
the Kongzi jiyu. The first step towards creating a purified corpus of authentic material is excluding the inauthentic material.
Early authors also recognized that Kongzi attributions could be motivated by something other
than a sincere desire to accurately represent Kongzi’s teachings. One early author in particular is
famous for having owned up to the fictionalized nature of his attributions:
The imputed words that are nine-tenths [of this text] make use of outside [figures] in their
discussions. The father does not act as his own son’s matchmaker—it is better for someone
who is not the father to praise [a potential wife] than for the father to do it himself. This is not
my fault, it is the fault of others. People respond to those who resemble themselves and they
turn away from those who differ. They approve of those who are similar and they blame those
who are different.
寓言十九，藉外論之。親父不為其子媒。親父譽之，不若非其父者也；非吾罪也，人之
罪也。與己同則應，不與己同則反；同於己為是之，異於己為非之。233
This passage from the “Yu yan 寓言” chapter of the Zhuangzi is the locus classicus for the term
“imputed words” (yu yan 寓言). The author of the “Xiu wu 修務” (“Striving to cultivate oneself”) chapter of the Huainanzi justified the practice of imputing words in similar fashion:
Vulgar men typically respect the ancient and belittle the new, and so when advocating
something one must attribute it to the Divine Farmer or Yellow Emperor before one can
venture a persuasion [or explanation]...Now if we were to take some new “writings of the
sages” and attribute them to Kongzi or Mozi, there would be many followers who would pick
over every phrase and accept the text [as authentic].
世俗之人，多尊古而賤今，故為道者必托之於神農、黃帝而後能入說。...今取新聖人
書，名之孔、墨，則弟子句指而受者必眾矣。234
Interestingly, both of these authors acknowledged that putting words in the mouths of famous
figures like Kongzi was a less than ideal form of writing. But they placed the ultimate responsibility for the practice on contemporary audiences who were too vulgar to appreciate texts that
233.See the “Tian xia 天下” (“All Under Heaven”) chapter of the Zhuangzi (SBCK 9/12a) for the explicit
association of Zhuang Zhou 莊周 with “imputed words.” The Shiji biography of Zhuang Zhou (Shiji 63.2143) also
uses the phrase in its description of Zhuang Zhou’s writings.
234.Huainanzi 19 (SBCK 19/11a). Mark Csikszentmihalyi (2004, p. 24) has identified this as “an ancient example of
what Michel Foucault called the ‘authorship function.’”
105
were not associated with legendary personas.
Other authors were far more critical of the practice of attributing words to figures for rhetorical purposes. One of these was Huan Tan 桓譚 (43 BCE-23 CE), who in his Xinlun 新論 (New
Discourses) charged Zhuangzi with taking the practice of “imputing words” to an extreme:
Zhuang Zhou imputed words, and so he said, ‘Yao asked Kongzi.’...This is why people often
say that his is a deficient and useless text.
莊周寓言，乃云堯問孔子...故世人多云短書，不可用。235
Huan Tan’s criticism is probably exaggerated. The received Zhuangzi contains no such anecdote,
nor do early texts preserve any examples in which Kongzi is placed in conversation with a legendary figure like Yao. (This is as ridiculous a scenario as a conversation between Abraham and
Jesus in the Biblical context or between Socrates and Julius Caesar in the Greco-Roman context.)
But Huan Tan might have singled out any number of Kongzi anecdotes in the Zhuangzi to make
the same point. Indeed, the Zhuangzi’s playful irreverence toward Kongzi marks a number of
episodes as blatantly fictionalized accounts, including the episode discussed above in which Yan
Hui ends up as Kongzi’s master (p. 97). Other examples include Kongzi’s meeting with the infamous “Dao Zhi 盜跖” (Robber Zhi) and his encounter with the sagely “Yu fu 漁夫” (“Fisherman”) in chapters 29 and 31, respectively. The former episode ends with Kongzi again asking to
be his interlocutor’s disciple,236 the latter with a chastened and shaken Kongzi admitting that he
had “barely escaped the tiger’s jaws” (幾不免虎口哉).237
At least one pro-Kongzi author also hinted that his Kongzi sayings were perhaps less than
perfectly authentic. Consider this passage from the Kong congzi, which tackles the Kongzi prob-
235.Quan Hou Han wen 13/2a, p. 537.
236.See Zhuangzi 31 (SBCK 10/5a–12a) and Zhuangzi 29 (SBCK 9/32b–47b).
237.Zhuangzi book 29 (SBCK 9/41a).
106
lem head-on in a dialogue between Zisi 子思, Kongzi’s grandson, and Duke Mu of Lu 魯穆公:
Duke Mu said to Zisi, “There are those who suspect that the Master’s sayings recorded in your
writings are actually your very own words.” Zisi replied, “Among the sayings of my
grandfather recorded in my writings are some which I have personally heard and some which
were brought to my attention by others. So even though my writings consist of words which
are not precisely the Master’s, they do not fall short of the Master’s ideas. What is it that you
doubt?” The Duke said, “The content [of your writings] is faultless.” Zisi said, “It has no
errors because it consists of my forefather’s ideas. But let us suppose that what you just said is
correct and that they are my own words. Since my words are faultless, they would still be
worthy of honor. But since it is not the case [that Kongzi’s words are really my own], why
doubt it?”
穆公謂子思曰：子之書所記夫子之言，或者以謂子之辭。子思曰：臣所記臣祖之言，或
親聞之者，有聞之於人者。雖非其正辭，然猶不失其意焉。且君之所疑者何？公曰：於
事無非。子思曰：無非所以得臣祖之意也。就如君言以為臣之辭，臣之辭無非，則亦所
宜貴矣。事既不然，又何疑焉？238
Zisi’s admission that some of his Kongzi attributions “are not precisely the Master’s” is striking,
as is his acknowledgement of the possibility that some quotations were actually his own words.
Also notice how Zisi hedges his bets by arguing that his Kongzi sayings are still “worthy of honor” (yi gui 宜貴) even if invented. Zisi is clearly unwilling to admit to “imputing words” outright
as the Zhuangzi author does. But the coy, half-hearted defense of his Kongzi sayings suggests
that the author of this dialogue, like the Zhuangzi author, was writing for an audience too sophisticated to accept just any Kongzi attribution as a record of the Master’s actual teachings.
Early authors’ occasionally explicit but typically implicit embrace of “imputed words” raises
the possibility, troubling for defenders of the traditional view, that early authors generally used
Kongzi as a puppet for their own views (or as a straw man for the views they wished to criticize).
Perhaps the Zhuangzi differs from other early texts only insofar as its “imputed words” flaunted
the conventions of mainstream Kongzi quotation practice by undermining Kongzi’s authority.
Perhaps early authors generally sought to represent Kongzi’s “ideas” (yi 意) instead of quoting
238.Kong congzi 9 (SBCK 3/5a). Translation adapted from Ariel 1989, p. 117.
107
his sayings verbatim. Of course, determining whether a given Kongzi saying was created by the
quoting author or copied more or less faithfully from some earlier source is usually impossible.
Most early authors were far less honest than the “Yu yan” author about their fictions, and even a
Kongzi saying that was originally fabricated might have been transmitted and quoted anew by
those who did not doubt its authenticity. Nevertheless, the patently fantastic or contrived nature
of some Kongzi sayings does allow some conjectures about the circumstances of their creation.
Consider the “Kongzi xianju 孔子閒居” chapter of the Liji (with close parallels in the Kongzi
jiayu and the “Min zhi fumu 民之父母” manuscript from the Shanghai Museum corpus), which I
discussed earlier as an example of a Kongzi commentary:
Kongzi was resting in leisure with Zixia in attendance when Zixia said, “May I ask about the
Shi [#251] that says ‘All happiness to our lord / Father and mother to the people?’ What sort of
person can be called ‘father and mother to the people?’”
Kongzi said, “The ‘father and mother of his people’ must have insight into the source of
ritual and music, and must achieve the ‘five reaches’ and practice the ‘three withouts.’ Thus
[his virtue] will pervade the world. When there is some loss within the four directions he will
always know it in advance. Such a man is referred to as ‘father and mother to the people.’”
Zixia said, “Now I have heard about the ‘father and mother to the people.’ May I ask
what the ‘five reaches’ refer to?”
Kongzi said, “Where the mind reaches the Shi reach, too; where the Shi reach ritual
reaches, too; where ritual reaches music reaches, too; where music reaches sorrow reaches,
too. Sorrow and joy give birth to each other. This is why when we look with clear eyes we
cannot see it, and when we cock our ears and listen for it we cannot hear it. When one’s mind
and qi fill heaven and earth, this is called the ‘five reaches.’”
Zixia said, “Now I have heard about the ‘five reaches.’ May I ask what the ‘three
withouts’ refer to?”
Kongzi said, “Music without sound, ritual without body, and mourning without dress—
these are the ‘three withouts.’”
Zixia said, “Now I have heard about the ‘three withouts.’ May I ask which Shi
approximate them?”
Kongzi said, “‘Day and night he buttressed the charge’ [Mao 271]—this is music without
sound; ‘I have borne myself correctly / In rites more than can be numbered’ [Mao 26]—this is
ritual without body; ‘When any of your people were in trouble / I crawled on my knees to help
them’ [Mao 35]—this is mourning without attire.”
Zixia said, “Your words are great and beautiful and magnificent.”
孔子閒居，子夏侍。子夏曰：敢問詩云：凱弟君子，民之父母。何如斯可謂民之父母
矣。孔子曰：夫民之父母乎，必達於禮樂之原，以致五至而行三無，以橫於天下。四方
有敗，必先知之。此之謂民之父母矣。子夏曰：民之父母，既得而聞之矣。敢問何謂五
至？孔子曰：志之所至，詩亦至焉；詩之所至，禮亦至焉；禮之所至，樂亦至焉；樂之
所至，哀亦至焉。哀樂相生，是故正明目而視之，不可得而見也；傾耳而聽之，不可得
108
而聞也。志氣塞乎天地，此之謂五至。子夏曰：五至既得而聞之矣。敢問何謂三無？孔
子曰：無聲之樂，無體之禮，無服之喪，此之謂三無。子夏曰：三無既得略而聞之矣。
敢問何詩近之？孔子曰：夙夜其命宥密，無聲之樂也。威儀逮逮，不可選也。無體之禮
也。凡民有喪，匍匐救之，無服之喪也。子夏曰：言則大矣美矣盛矣。239
“Kongzi xian ju” reads as a Kongzi yue commentary in the form of a conversation between
Kongzi and Zixia. Earlier I labeled it a “clarifying” commentary insofar as it provides a specific
frame of reference—li 禮 (ritual) and yue 樂 (music)—for a Shi couplet which is otherwise open
to interpretation. Zixia’s follow-up questions also mark Kongzi’s initial statement as an object of
commentary in its own right. “What do you mean?” (he wei ye 何謂也) is the only natural response to Kongzi’s odd notions of the “five reaches” (wu zhi 五至) and “three withouts” (san wu
三無), odd because they turn two words typically used as verbs—zhi 至 (to reach) and wu 無 (to
not have)—into countable nouns.240 The “five reaches” and “three withouts” seem to have been
designed in order to prompt explanation.
Beyond this, there is something rather artificial about the presentation of the dialogue, beginning with its emphatic and repetitive section-marking. Kongzi’s statements conclude with the
phrase “this is what X means,” which makes Zixia’s interjections (“Now I have heard about X.
May I ask about Y?”) segueing from the end of one section to the beginning of the next entirely
redundant. I take this as a sign that the anecdote was designed to be something other than a realistic depiction of an actual conversation.
Suppose that we excised all of Zixia’s questions from the anecdote—would that impair our
understanding of the text? No, because the repetition of the phrase “This is what X means” al-
239.Liji 29 (SBCK 15/9b–11a; KZJY pp. 545–6). See also Shanghai bowuguan cang Zhanguo chu zhushu vol. 2, pp.
154ff and Kongzi jiayu 27 (SBCK 6/17b–18b). For present purposes, I set aside a number of significant differences
among these versions. Shi translations from Waley 1996. For other instances of “enumeration” in Warring States
rhetoric, see Schaberg 2001, p. 47.
240.See Lunyu 20/2 for a similarly structured dialogue that unpacks the concepts of the wu mei 五美 (five
excellences) and si e 四惡 (four evils).
109
ready signals the transitions in Kongzi’s commentary. Let us go one step further and remove the
Kongzi yue markers while keeping the opening Shi quotation—what remains? The answer, I
think, is a perfectly well-structured piece of Warring States rhetoric:
A Shi says, “All happiness to our lord / Father and mother of his people?” The “father and
mother of his people” must have insight into the source of ritual and music, and must achieve
the five reaches and practice the three withouts. Thus [his virtue] will pervade the world.
When there is some loss within the four directions he will always know it in advance. Such a
man is referred to as “father and mother of his people.”
Where the mind reaches the Shi reach, too; where the Shi reach ritual reaches, too; where
ritual reaches music reaches, too; where music reaches sorrow reaches, too. Sorrow and joy
give birth to each other. This is why when we look with clear eyes we cannot see it, and when
we cock our ears and listen for it we cannot hear it. When one’s mind and qi fill heaven and
earth, this is called the “five reaches.”
Music without sound, ritual without body, and mourning without dress—these are the
“three withouts.”
“Day and night he buttressed the charge”—this is “music without sound”; “I have borne
myself correctly / In rites more than can be numbered”—this is “ritual without body”; “When
any of your people were in trouble / I crawled on my knees to help them”—this is “mourning
without attire.”
詩云：凱弟君子，民之父母。夫民之父母乎，必達於禮樂之原，以致五至而行三無，以
橫於天下。四方有敗，必先知之，此之謂民之父母矣。志之所至，詩亦至焉；詩之所
至，禮亦至焉；禮之所至，樂亦至焉；樂之所至，哀亦至焉，哀樂相生。是故正明目而
視之，不可得而見也；傾耳而聽之，不可得而聞也。志氣塞乎天地，此之謂五至。無聲
之樂，無體之禮，無服之喪，此之謂三無。夙夜其命宥密，無聲之樂也。威儀逮逮。不
可選也，無體之禮也。凡民有喪，匍匐救之，無服之喪也。
In this reading, Zixia’s questions are artificial because they were motivated primarily by pedagogical concerns, with the dialogical, catechistic framework facilitating the analysis and memorization of the material. If we ask what the ultimate lesson of “Kongzi xian ju” might have been,
one (admittedly speculative) possibility is that it was intended as teaching tool for composing
well-structured speeches out of quotations from canonical texts. The rhetorical template would
run as follows: open with a quotation from a Shi; use that quotation to introduce one’s main ideas
(in this case, the “five reaches” and “three withouts”); for every main idea, add another section
which expands upon that idea; finally, conclude with a canonical quotation or two (or, in this
case, three) to show that one’s conclusions are consistent with the canon.
110
In any event, my suspicion is that the intended audience of “Min zhi fumu” would have understood Kongzi yue as a marker of the text’s pedagogical function without also seeing it as a
mark of authenticity.241 This interpretation also casts the framing of the passage—Kongzi xian ju
孔子閒居 (Kongzi was resting in leisure), a formula which introduces similar dialogues in the
Liji, Kongzi jiayu, Hanshi waizhuan, and Shuiyuan—in a new light.242 Prefacing a Kongzi dialogue in this way might have signaled that the discussion was not prompted not by a specific occasion or political context as in “Lu bang da han 魯邦大旱.” The formula’s apparent allusion to
the “Bei shan 北山” (“Northern Hills,” #205) ode—“Some people sit quietly at home, / Others
wear themselves out in serving their state” (或燕燕居息，或盡瘁事國)—would seem to support
this reading.243 In keeping with the image of “sitting quietly at home,” the Kongzi of these anecdotes addresses topics of general interest without engaging contemporary political affairs. However, the absence of the Kongzi xian ju 孔子閒居 line in the “Min zhi fumu” manuscript might
suggest that it developed in a later period as a way of marking and categorizing this sub-genre of
Kongzi dialogue.
The use of Kongzi disciples as stand-ins for certain themes or concerns is another hint that at
least some early authors thought of disciples more as convenient props or symbols than as actual
historical figures, e.g., in this Kongzi comment from the Yanzi chunqiu 晏子春秋 which follows
an anecdote about Yanzi’s frugal mourning rituals for his deceased father:
241.Weingarten (2010, p. 140) makes a parallel argument about a sub-genre of Kongzi dialogue revolving around
water and mountains: “Confucius and one of his disciples act as dramatis personae; they flesh out the purely textual
functions of the preceding versions.”
242.See the “Zhongni yan ju 仲尼燕居” chapter of the Liji, chapter 27 of the Kongzi jiayu, chapter 37 of the Da Dai
liji, and Hanshi waizhuan 7 (SBCK 7/14b–15a; KZJY 8.6/pp. 160–61) and 8 (SBCK 8/13a–b; KZJY 1.8/p. 4). All of
these variants (e.g., yan ju 燕居 and xian ju 閒居) rhyme on *-en.
243.SBCK 13/7a. See also Zheng Xuan’s gloss on the title of the “Kongzi xianju” chapter of the Liji: “retiring and
avoiding others is called ‘resting in leisure’” (退燕避人曰閒居).
111
When Zengzi asked Kongzi about it, Kongzi said, “Yanzi can be said to have been ‘able to
keep harm far from himself.’ He did not rebuff another’s error with his own correctness but
avoided blame through deferential speech. What propriety!”
曾子以聞孔子，孔子曰：晏子可謂能遠害矣。不以己之是駁人之非，遜辭以避咎，義也
夫！244
The Kongzi comment is typical in that it suggests a tagline for the virtue displayed by Yanzi in
the anecdote—neng yuan hai 能遠害 (being able to keep harms at a distance)—and evaluates
Yanzi’s actions according to some abstract principle, here “propriety” (yi 義). As we have seen,
many Kongzi comments are framed as conversations with disciples. But Zengzi’s mention in this
instance is probably not accidental. As attested in the first section of the Xiaojing, among other
texts,245 Zengzi was more closely associated with the value of xiao 孝 (filial piety) than any other
Kongzi disciple. Mentioning Zengzi in conjunction with the Kongzi comment was a way of
signaling that the episode could be categorized under the heading “filial piety” or “relations between fathers and sons.” Tellingly, Zengzi himself does not even need to speak to fulfill this role.
His mere mention is enough to mark the comment in a certain way.
Scattered passages in the Mengzi, Lunyu, and elsewhere indicate that Zengzi was not the only
figure to be thought of in symbolic terms:
<Lüshi chunqiu 17/7>Lao Dan prized pliancy; Kongzi prized humaneness; Mo Di prized
integrity; Guan Yin prized purity; Zi Liezi prized emptiness; Chen Bing prized equanimity;
Yang Sheng prized himself; Sun Bin prized strategic positioning; Wang Liao prized being
first; Ni Liang prized being last.
老耽貴柔，孔子貴仁，墨翟貴廉，關尹貴清，子列子貴虛，陳駢貴齊，陽生貴己，孫臏
貴勢，王廖貴先，兒良貴後。246
244.KZJY 9.29/p. 184. Translation adapted from Schaberg 2005, p. 16.
245.Zengzi is also the most widely quoted authority in the “Xiao xing 孝行” (“Filial conduct”) chapter of the Lüshi
chunqiu (14/1; SBCK 7/5b ff) and the “Zengzi ben xiao 曾子本孝” (“Zengzi makes filial piety the root”), “Zengzi li
xiao 曾子立孝” (“Zengzi establishes filial piety”), “Zengzi da xiao 曾子大孝” (“Zengzi aggrandizes filial piety”),
and “Zengzi shi fumu 曾子事父母” (“Zengzi served his father and mother”) chapters of the Da Dai Liji (books 50–
53).
246.SBCK 17/18a; KZJY 14.39/p. 316.
112
<Mengzi 2A/2>“Zai Wo and Zigong are good at persuasion and fine phrases. Ran Niu, Minzi,
and Yan Yuan are good at speaking about virtuous conduct.”
宰我、子貢善為說辭；冉牛、閔子、顏淵善言德行。247
<Lunyu 11/3>Virtuous conduct: Yan Yuan, Min Ziqian, Ran Boniu, Zhonggong. Speaking:
Zai Wo, Zigong. Government affairs: Ran You, Ji Lu [probably Zilu]. Refinement and
learning: Ziyou, Zixia.
德行：顏淵、閔子騫、冉伯牛、仲弓。言語：宰我、子貢。政事：冉有、季路。文學：
子游、子夏。
There are at least two ways of interpreting such lists, most obviously as descriptions of these figures’ primary virtues as they emerge in early anecdotal literature. From this perspective, the
Lüshi chunqiu author was simply noting that Kongzi often spoke about “humaneness” (ren 仁),
the Lunyu 11/3 author that Yan Yuan was often represented as a paragon of virtue, etc. But we
might also read them as suggestions or even prescriptions for generating new sayings attributed
to Kongzi, Kongzi disciples, or other masters.248 An author who wanted to buttress his argument
about “pliancy” (rou 柔) with an appeal to authority should find or create a relevant Laozi saying. Someone who wanted to create a Kongzi dialogue on the theme of “virtuous conduct”
should cast Yan Hui (or Min Ziqian or Ran Boniu or Zhonggong) as Kongzi’s interlocutor. And
so on.
Puns on disciple names are another indication that Kongzi disciples were thought of in symbolic, not historical, terms. Consider the following examples from the Lunyu:249
<Lunyu 2/9>The Master said, “I can talk to Yan Hui [*ɢwʕəәj] all day long without him
disagreeing [*ɢwəәj] with me, as if he was stupid...”
子曰：吾與回言終日，不違，如愚。...
<Lunyu 5/4>Zigong asked, “What am I like?” The Master said, “You are a vessel.” “What
247.SBCK 3/8b; KZJY p. 510.
248.A clear example of this possibility in the Lunyu 12/1. For a discussion, see chapter three, pp. 245–249.
249.Some of these puns are pointed out in Brooks & Brooks 1998, pp. 22–23.
113
kind of vessel?” “A jade vessel.”
子貢問曰：賜也何如？子曰：女、器也。曰：何器也？曰：瑚璉也。
<Lunyu 6/7>The Master said, “Yan Hui [*ɢwʕəәj] can go three months without disobeying
[*ɢwəәj] in his heart. The rest of them only manage it for the occasional day or month.”
子曰：回也，其心三月不違仁，其餘則日月至焉而已矣。
<Lunyu 9/12>When the Master was gravely ill Zilu sent his retainers to serve him as a
subjects. After his illness had lessened Kongzi said, “Zilu [*Cəә.rʕak-s)] has been deceitful
[*tsʕrak-s] for some time now! Who am I fooling by [pretending] to have subjects although I
have none? Am I fooling heaven? Wouldn’t I rather die under the care of my disciples than
these subjects? And even if I cannot have a great funeral, will I die along some road
[*Cəә.rʕak-s)]?”
子疾病，子路使門人為臣。病間，曰：久矣哉，由之行詐也！無臣而為有臣。吾誰欺？
欺天乎！且予與其死於臣之手也，無寧死於二三子之手乎！且予縱不得大葬，予死於道
路乎！
<Lunyu 19/25>Chen Ziqin said to Zigong [*kʕoŋ-s], “You are reverential [*k<r>oŋ] indeed—
how could Zhongni be more accomplished than you?”
陳子禽謂子貢曰：子為恭也，仲尼豈賢於子乎？
Both Lunyu 2/9 and 6/7 juxtapose the “Hui 回” of Yan Hui with the homophone wei 違 (disagree
or disobey), just as Lunyu 19/25 juxtposes the “gong 貢” of Zigong with gong 恭 (reverential).
Lunyu 9/12 not only rhymes “Zilu” with “deceitful” (zha 詐), it also echoes Zilu’s name in the final word, “road” (dao lu 道路). To an early Chinese ear, the last line might have even read as
“Will I die leading [Zi]lu?” or perhaps “Will I die speaking to [Zi]lu?” Finally, Lunyu 5/4 plays
on the fact that Zigong’s style name includes the word gong 貢, meaning “to offer [i.e., as in a
sacrifice],” and describes him as a precious sacrificial vessel. Of course, there is nothing to preclude the possibility that the historical Kongzi himself enjoyed the occasional pun, as Cristoph
Harbsmeier has suggested.250 But we should also consider the possibility that the frequent appearance of specific disciples in early Kongzi dialogues and anecdotes reflects neither the origi-
250.On humor in the Lunyu, see Harbsmeier 1990.
114
nal circumstances of Kongzi’s instruction nor the social context of a later intellectual lineage (jia
家). Rather, the rotating cast of colorful characters may be nothing more than a set of literary
conventions for marking, categorizing, and enlivening early Kongzi sayings.
How did early authors conceive of Kongzi quotation practice?
I have already discussed several passages in which early authors commented on the practice of
Kongzi quotation. Youzi’s criticism of Zengzi’s Kongzi quotation in “Tan gong” showed that
Kongzi sayings were subject to critical standards; the “man of Chu lost his bow” episode indicated that Kongzi (and Laozi) sayings were designed to be maximally gong 公 (general or abstract);
the “Yu yan” chapter of the Zhuangzi called attention to the practice of imputing words to famous figures (p. 104); a passage from the Kong Congzi described Kongzi’s Shu commentary as
having explained the “greater meaning” of the Shu (p. 74); and Wang Chong wondered how
Kongzi’s words at Lunyu 14/25 could be so “sparing” (yue 約) and opaque (p. 88). Such examples do not begin to exhaust early authors’ reflections on Kongzi and the sayings attributed to
him.
The following passage from the Mozi, the most stridently anti-Kongzi text from the early period, ironically provides the clearest demonstration of Kongzi sayings’ importance within early
intellectual discourse:
Master Mo was engaged in disputation with Master Cheng when he cited Kongzi. Master
Cheng asked him, “How can you criticize the Ru and cite Kongzi?” Master Mo said, “This is a
case of something being both appropriate and unalterable. Now when birds learn of vexing
heat and drought they fly up high, and when fish learn of vexing heat and drought they swim
downward. In situations like these even the best-laid plans of Yu and Tang could not alter this.
Although birds and fish can be called foolish, even Yu and Tang would follow them at times.
Now should I never cite Kongzi?
子墨子與程子辯，稱於孔子。程子曰：“非儒，何故稱於孔子也？”子墨子曰：“是亦當
115
而不可易者也。今鳥聞熱旱之憂則高，魚聞熱旱之憂則下，當此雖禹湯為之謀，必不能
易矣。鳥魚可謂愚矣，禹湯猶云251因焉。今翟曾無稱於孔子乎？”252
Although he elsewhere accuses Kongzi of being a rank hypocrite (p. 96) and a purveyor of banal
wisdom (p. 96), here Mozi nevertheless acknowledges the “appropriateness” (dang 當), and “unalterability” (bu ke yi 不可易) of quoting Kongzi. Kongzi sayings were apparently so ubiquitous
that not even Kongzi’s detractors could do entirely without them.253
Authors who, like Youzi in “Tan gong,” sought to police the corpus of Kongzi sayings and
defend Kongzi against spurious quotations are particularly fruitful sources for early quotation
practice. The Kong congzi includes two episodes in which Kongzi’s disciple Zigao 子高 debunks
rumors about his master, in the first that Kongzi was a heavy drinker and in the second that he
met inappropriately with two women (Appendix 1:D). Zigao’s strategy for dealing with these
“inciting and provocative words” (勸厲獎戲之辭) and “gossip” (流言) is instructive. He begins
by insisting that, as a sage, Kongzi must have acted sagely; ergo, he could not have been a drunk:
“it was by means of the Way and virtue that worthies and sages surpassed others, not through
food and drink” (賢聖以道德兼人，未聞以飲食也). In the second instance, he simply asserts
that Kongzi was not someone who “did not grasp [the rites]” (夫子亦弗獲巳矣), thus he would
not have violated ritual propriety in order to meet with two women. Zigao also draws an ethical
contrast between himself and those who “love to drink” (嗜酒者), who use Kongzi sayings to
“incite and provoke” (勸厲), and who “fake [stories] in order to further their own motives” (假其
類以行其心者). In the second example, Zigao then presents an alternate history of Kongzi’s ad-
251.Understanding yun 云 as huo 或. See Mozi jiangu p. 278.
252.Mozi book 48 (SBCK 12/13b–14a)
253.Compare Denecke 2004, p. 154: “The attempted belittling of Confucius cannot conceal that Mozi recognizes his
words as common sense.”
116
ventures in Wei which leaves no room for an improper meeting with the Duke of Wei’s wife.
Taken together, Zigao’s critiques provide a sense of what was required to maintain Kongzi’s integrity and the integrity of Kongzi quotation practice: consensus about Kongzi’s status as a
“sage” (sheng 聖), agreement about the details of Kongzi’s biography, and a shared commitment
not to “fake” (jia 假) new stories or sayings. If the Kong congzi is any indication, the effort to
defend Kongzi quotation practice may have been an uphill battle.
The author of Mengzi 5A/8 encountered a similar problem as Zigao in the Kong congzi (Appendix 1:E) and also adopted a similar strategy for refuting spurious rumors: attack those who
fabricated the stories, present an alternate biography, and appeal to the image of Kongzi as the
kind of figure who, by definition, would not have compromised his integrity. The phrasing of the
rhetorical question which concludes the episode—“If Kongzi had stayed with an ulcer doctor and
the servant Qi Huan, how could he have been Kongzi?”—shows which of these arguments carried the most weight.254 Mengzi simply could not entertain the possibility that Kongzi was less
than perfectly virtuous.255 In the end, the demands of the Kongzi brand trumped all else.
A similar exchange takes place between Mengzi and Xian Qiumeng 咸丘蒙 at Mengzi 5A/4:
Xian Qiumeng asked, “A saying has it that ‘a lord cannot make a minister out of a man of
flourishing virtue, nor can a father make him a son.’ Shun stood facing south and Yao led all
the vassal lords to face north at court. Shun’s father Gu Sou also faced north at court. When
Shun saw Gu Sou he furrowed his brow. Kongzi said, ‘At that time the world was endangered
and teetering on the edge.’ I do not know if this saying is true or not.”
Mengzi said, “No, this is not the saying of a noble man. It is the saying of a rube from
eastern Qi. When Yao was an old man Shun took over the government. The Canon of Yao
states that ‘after 28 years the Meritorious One [i.e., Yao] passed away, and the people grieved
as if they had lost a parent, and all within the four seas gave up music for a time.’ Kongzi said,
254.The question he yi wei Kongzi 何以為孔子 (How could he have been Kongzi?) is echoed in the Yantie lun by
one of the Ru 儒 debaters who counters the criticism that the Ru are neither junzi nor“worthy man of service” (xian
shi 賢士). Yantie lun 59 (SBCK 10/16a).
255.This recalls Kongzi’s defense of the King Wen from the Zhuangzi (p. 62): “King Wen was perfect, so how can
you evaluate and criticize him? He only used [the dream] to handle that particular situation” (夫文王盡之也，而又
何論刺焉).
117
“Heaven does not have two suns; the people do not have two kings.’ If Shun was already the
Son of Heaven and also led all the vassal lords to mourn Yao for three years, there would have
been two Sons of Heaven.”
咸丘蒙問曰：語云，盛德之士，君不得而臣，父不得而子。舜南面而立，堯帥諸侯北面
而朝之，瞽瞍亦北面而朝之。舜見瞽瞍，其容有蹙。孔子曰：於斯時也，天下殆哉，岌
岌乎！256不識此語誠然乎哉。孟子曰：否；此非君子之言，齊東野人之語也。堯老而舜
攝也。堯典曰：二十有八載，放勛乃徂落，百姓如喪考妣，三年，四海遏密八音。孔子
曰：天無二日，民無二王。舜既為天子矣，又帥天下諸侯以為堯三年喪，是二天子
矣。257
In the continuation of the dialogue, Xian Qiumeng quotes the “Bei shan 北山” (“Northern
Hills”) ode to ask Mengzi whether Shun’s father served Shun as his subject, the implication being that Shun’s position as emperor compromised his filial piety and vice versa.258 In his defense
of Shun, Mengzi refutes Xian Qiumeng’s reading (“This is not what this Shi means” [是詩也，
非是之謂也]) and goes on to establish some guidelines for Shi interpretation generally: “those
who would explain a Shi should not use the [interpretation of its] words to impair the [interpretation of its] phrasing, or the [interpretation of its] phrasing to impair the [interpretation of its] intent” (說詩者，不以文害辭，不以辭害志。).
The contrast between Mengzi’s handling of Xian Qiumeng’s Kongzi attribution and his handling of the Shi quotation is instructive. Xian Qiumeng’s mistake with the Shi is one of misinterpretation, of misconstruing a text which both men acknowledge is legitimate. Since Mengzi cannot reject the quotation itself, he must contradict Xian Qiumeng’s understanding of it. In
contrast, when Mengzi disagrees with a Kongzi saying (“At that time the world was at risk and
teetering on the edge”) he rejects it outright (“this is not the saying of a junzi”) and replaces it
with an entirely different saying that better supports his argument (“Heaven does not have two
256.A variant of this saying appears at Han Feizi 51 (SBCK 20/1b; KZJY ), where it is also criticized.
257.SBCK 9/6b–8a; KZJY pp. 513–14.
258.The quotation reads: “Everywhere under Heaven / Is no land that is not the king’s. / To the borders of all those
lands / None but is the king’s slave” (普天之下，莫非王土；率土之濱，莫非王臣。). See SBCK 13/6b.
118
sons; the people do not have two kings”). Herein lies one of the key differences between Shi and
Kongzi quotation practice in the early period, at least until the Lunyu was established as the preeminent source of Kongzi sayings. In the absence of a clearly delineated set of authoritative
Kongzi sayings, authors had to defend their own Kongzis and dispute others’ Kongzis as the situation demanded.
In this respect, the challenges of Kongzi quotation practice reflected a more general anxiety
in the period about the integrity of received textual traditions. Here we might recall Mengzi’s advice at Mengzi 7B/3: “Mengzi said, “It is better to go without Shu than to trust them completely.
I only accept two or three strips out of [the Shu text] ‘Martial Virtue Perfected’” (盡信書，則不
如無書。吾於武成，取二三策而已矣。). The Lüshi chunqiu compilers expressed a similar
sentiment when they urged others not to accept “traditions” (zhuan 傳) uncritically:
Received sayings must be scrutinized. After several transmissions white turns into black and
black turns into white. A dog looks like an ape, an ape looks like a monkey, and a monkey
looks like a person, but there is a huge difference between a dog and a man. This is how stupid
people make big mistakes. To examine what one hears leads to good fortune. But it would be
better not to have heard something than to fail to examine what one has heard.
夫得言不可以不察，數傳而白為黑，黑為白。故狗似玃，玃似母猴，母猴似人，人之與
狗則遠矣。此愚者之所以大過也。聞而審則為福矣，聞而不審，不若無聞矣。259
The authors quoted in this section clearly took the Lüshi chunqiu’s advice to heart. According to
the traditional view of the Lunyu, Kongzi’s teachings were “recorded” (ji 記) and “edited and
compiled” (ji er lun zuan 輯而論篹) by Kongzi’s students before being transmitted faithfully by
generations of devoted disciples. But the examples considered here reveal that the early authors
who championed Kongzi did not simply transmit the Master’s words. Sometimes they encountered sayings that contradicted their own views of Kongzi, and disagreements over Kongzi attri-
259.Lüshi chunqiu 22/6 (SBCK 22/10b). I follow Knoblock & Riegel (2000, p. 582) in emending the first de 得
(obtain) to zhuan 傳 (transmitted or received).
119
butions forced them to adopt a more critical approach to Kongzi traditions.260 In the apparent absence of a Kongzi canon, they developed strategies for limiting spurious attributions, including
by filling in the details of Kongzi’s biography or insisting on Kongzi’s exemplary status. Not
until the advent of the Lunyu as an authoritative collection of Kongzi’s teachings were proKongzi authors freed from having to actively police Kongzi’s reputation.
Conclusions
Even after more than one-hundred pages and several dozen Kongzi sayings, this introduction has
only scratched the surface of the topic. Kongzi quotation practice in the early period was a decidedly messy phenomenon, so much so that one might reasonably wonder whether Kongzi was
nothing more than a cipher who parroted whatever ideas an author wished to attribute to him.
But while Kongzi yue sayings were certainly a flexible medium, they were not infinitely flexible.
Close attention to the functions of Kongzi sayings reveals a handful of common threads and interests which, taken together, illuminate their distinctive role within early intellectual discourse:
(1) Kongzi sayings are highlighters. This is one feature of Kongzi sayings that sets them
apart from other kinds of quotable material. Dependent Kongzi sayings focused attention on texts
their authors deemed worthy of comment, be they historical anecdotes, other sayings, or canonical texts. Especially in large compilations like the Zuozhuan and Lüshi chunqiu, Kongzi comments singled out particular episodes from among hundreds of others, thus ensuring that this material received special consideration from early audiences.
260.For another example, see Zhuangzi book 27 (SBCK 9/13b; KZJY 13.20/pp. 280–81) and Xunzi book 8 (SBCK
4/12b).
120
(2) Kongzi sayings “transmit without originating.” This is another feature that distinguishes
Kongzi sayings from other quotable material. Kongzi sayings frequently quoted other sources of
received wisdom, from the anonymous “mottos” (zhi 志) quoted by Zuozhuan Kongzi comments
(p. 55) to the Shi quotations of “Min zhi fumu” (p. 108) or the nursery rhymes of the Shuiyuan
(p. 67). This aspect of Kongzi quotation practice is perhaps best captured by the Lunyu 7/1
phrase, “transmitting without originating” (shu er bu zuo 述而不作). Although intended in the
Lunyu as a description of Kongzi’s reverence for the ancients, it is also an apt characterization of
the function of Kongzi sayings in early textual culture: invoking and adapting received wisdom
to contemporary concerns without necessarily introducing new ideas of their own.
(3) Kongzi sayings are venues. An even more accurate description of Kongzi sayings might
be “transmit and create” (shu er zuo 述而作), because some Kongzi sayings did much more than
simply highlight and transmit other texts. Kongzi sayings were also a dynamic genre through
which early authors reworked earlier texts and maximized their rhetorical potential. Kongzi yue
could signal a performance of rhetorical artistry, an occasion for introducing a bit of “patterning”
(wen 文) to make words “go far” (yuan 遠), to paraphrase a Kongzi saying from the Zuozhuan
(p. 55).
(4) Kongzi sayings are verbal insignia. Han authors in particular prized Kongzi’s role as the
source of the “greater meaning” (da yi 大義) of the canon, i.e., the original, pristine wisdom prior to its corruption in the Warring States and Qin periods (p. 74). In practice, the “greater
meaning” was not a commentary so much as a handy tagline or motto which captured a text’s
core lesson, that controlled its interpretation, and that fixed its place in the early Chinese textual
universe. The precursors of “greater meaning” statements were the emblematic ideas or phrases
that Kongzi comments affixed to specific anecdotes and figures in Warring States anecdotal
121
literature.
(5) Kongzi sayings are teaching texts. The pedagogical motivation behind many Kongzi sayings is apparent in numerous Kongzi yue comments and commentaries, from the way that Kongzi
modeled a direct, personal connection with canonical learning (p. 77), and in the use of a Kongzi
dialogue as a rhetorical exercise (p. 108). The framing of so many Kongzi sayings as scenes of
instruction involving Kongzi’s disciples or other interlocutors likely also reflects a pedagogical
context.
(6) Kongzi sayings promote the canon. This is true in a concrete sense insofar as Kongzi yue
commentaries on the Shi, Shu, and Yi trumpet the value of the canon. But Kongzi sayings also
demonstrate that value by abstracting lessons from specific canonical texts.261 In his study of early canonical commentaries, John Henderson identified six fundamental assumptions about
canonicity from a number of ancient textual traditions: the canon is supposed to be “comprehensive,” “well ordered and coherent,” “self-consistent,” “moral,” “profound,” and “clear and accessible.”262 One can find all of these assumptions at work in the corpus of Kongzi sayings. As we
have seen, the Kongzi yue meta-commentaries of the “Kongzi shilun” and Zhouyi (p. 82) insist
on the clarity and accessibility of the Shi and Yi; the amplifying commentary of the “Kongzi
shilun” demonstrates the profundity of the Shi (p. 72); the “Wen yan” provides an ethical frame
of reference which makes the enigmatic line statements of the Zhouyi accessible (p. 78); and the
self-consistency, coherence, and comprehensiveness of, e.g., the various royal speeches that
make up the early Shu collection are demonstrated in the “greater meaning” Kongzi commentary
261.One fine example of this tendency is a dialogue between Kongzi and Zigong from Xunzi book 27 (SBCK
19/20a–b; KZJY 1.9/pp. 4–5) that features Kongzi extracting lessons from various Shi couplets.
262. See Henderson 1991, pp. 100–133.
122
from the Kong congzi (p. 74), which pigeonholes the speeches within a matrix of traditional
virtues.
One way to characterize the relationship between Kongzi sayings and canonical traditions is
with a term borrowed from the title of the Han Shi waizhuan 韓詩外傳 (Outward Commentary
on Mr. Han’s Odes), a compilation of anecdotes capped with quotations from the Shi.263 Instead
of explicating a source text, an “outward commentary” (wai zhuan 外傳) demonstrates its didactic and rhetorical potential by applying its quotations to superficially unrelated material. Considered as a whole, the corpus of Kongzi sayings might also be described as an “outward commentary” of sorts insofar as Kongzi sayings often demonstrate the quotability of received textual
traditions in various contexts. The myriad intersections between Kongzi yue sayings and canonical traditions may have also helped to create the impression that these disparate texts formed a
single, internally coherent tradition anchored to Kongzi. Kongzi sayings were among threads that
bound the tradition together. Little wonder, then, that Han authors looked to Kongzi as the creator of the Classics.
(7) Kongzi sayings are prompts. The independent Kongzi sayings I have labeled “Kongzi
sayings as objects of commentary” functioned very differently from dependent Kongzi sayings.
These brief, enigmatic nuggets of wisdom demand explanation in ways that dependent sayings
do not, and thus reflect the perception that a sage like Kongzi needed to sound sagely. Unlike
other sayings, prompt sayings seem to have been consciously designed to seem profound.
(8) Kongzi sayings are shared intellectual property. It comes as no surprise that the Mengzi
263.Hightower’s (1952) translation of the title—Han Ying’s Illustrations of the Didactic Application of the Classic
of Songs—makes the same basic point. See also Hightower’s entry on the Hanshi waizhuan in Loewe (ed.) 1993, p.
125.
123
孟子, Xunzi 荀子, and other texts retrospectively labeled “Ru 儒” frequently quoted Kongzi,
whom they upheld as a sage. More surprising is the fact that intellectual rivals of the Ru, i.e., the
thinkers retrospectively labeled as “legalist” (fajia 法家), “Mohist” (mojia 墨家), and “daoist”
(daojia 道家), also made frequent use of Kongzi sayings.264 Even in the Mozi, the Warring States
text most critical of Kongzi and his teachings, we find Mozi himself defending his appeal to
Kongzi as a source of quotable wisdom (p. 115).265 Quoting Kongzi was clearly not an esoteric
practice.
*
*
*
Let us also dwell for a moment on what my survey has left out. Unlike Sun Xingyan, who
grouped Kongzi sayings in the Kongzi jiyu into fourteen chapters based largely (although not exclusively) on their content, I have not built philosophical or thematic considerations into my typology. Was this a mistake?266
Earlier I established a distinction between dependent sayings—sayings that exist in order to
comment on or elaborate other texts—and independent sayings—those that present ideas worthy
of consideration in their own right. An example of a dependent saying was the Kongzi comment
on the Ci Fei anecdote in the Lüshi chunqiu which praises Ci Fei with language lifted directly
from the anecdote itself (p. 48). My reluctance to focus on Kongzi’s philosophy or doxography
stems from my inability to discern any philosophical content whatsoever in the Ci Fei comment
and similar sayings. In such instances, authors did not quote Kongzi for the content of his teach-
264.See Nylan & Csikszentmihalyi 2003 for a critique of this taxonomy, whose earliest expression is Shiji
130.3288–3292. For a useful counterpoint, see Klein 2010, p. 45, n. 1.
265.The antagonists of the Ru depicted in the Yantie lun 鹽鐵論 (Salt and Iron Discussions), probably a late
Western Han text, also regularly cited Kongzi to buttress their arguments.
266.For the argument that philosophy is crucial to understanding the genesis of the Lunyu, see Slingerland 2000 and
Goldin 2011.
124
ings.267 A Kongzi yue comment was a way to mark, to evaluate, and to expand upon another text.
Here we might recall the “man of Chu lost his bow” episode from the Lüshi chunqiu which expressed the difference between Kongzi and Laozi primarily in terms of competing rhetorical
strategies. Kongzi sayings deal with human universals and Laozi sayings with cosmological universals, and an early author could invoke either figure depending on the degree of “impartiality”
(gong 公) or abstractness he was looking to express. To be clear, I do not deny that the ideas of
Kongzi sayings mattered to Kongzi quoters. But demanding “philosophy” (however we define it)
from Kongzi sayings prejudges the corpus by assigning uninformative sayings a secondary
status.
What of Kongzi himself? Has my reluctance to tackle Kongzi biography disadvantaged my
survey in any way?
Early Chinese texts preserve any number of Kongzis, not all of whom are familiar from the
perspective of the later tradition.268 The challenge facing those who would argue that early texts
preserve at least some reliable information about the historical Kongzi is to determine which of
these myriad Kongzis is the real one. Certainly nothing in this chapter rules out the possibility.
But I hope to have demonstrated here that we need not commit ourselves to recovering the historical Kongzi in order to appreciate why early authors were so interested in quoting Kongzi.
Kongzi sayings were valuable rhetorical resources in their own right irrespective of their relationship to the historical Kongzi.
267.Denecke 2004, p. 143: “...Confucius did not matter so much for his arguments, but for his personal stature. The
vast corpus of lore about the master seems more interested in rendering his charisma as a person rather than in
transmitting a circumscribed and unified body of his teachings.”
268.Compare Gu Jiegang’s statement (1982, p. 130) that “each age has its very own Kongzi, and in each age there
are all sorts of different Kongzis” (各時代有各時代的孔子，即在一個時代中也有種種不同的孔子呢). For the
same point see Nylan & Wilson 2010, pp. 25–26.
125
On the other hand, even a Kongzi agnostic like myself has to acknowledge that many early
authors professed a belief in Kongzi’s historicity and in the accessibility of his teachings. One of
these authors was Sima Qian, who described his personal view of Kongzi in a comment on the
“Kongzi shijia”:
The Grand Scribe says, “There is a Shi that says, ‘I gaze up at the lofty mountain, / I travel the
great road.’ Although I cannot reach him, my heart goes towards him. When I read the
writings of Master Kong, I long to see him as a man. When I went to Lu, I saw Zhongni’s
temple, carriage, clothes, and ritual vessels, and various masters frequented his house
practicing rituals. I left there only with great reluctance. Many are the lords and accomplished
men who were renowned in their time yet died [and were forgotten]. Kongzi was a common
man who for more than ten generations has been revered by scholars. From the Son of Heaven
and the enfeoffed kings and lords, all those in the central states who speak of the Six Arts take
Kongzi as their guide. He can be said to be a perfect sage.”
太史公曰：詩有之：高山仰止，景行行止。雖不能至，然心鄉往之。余讀孔氏書，想見
其為人。適魯，觀仲尼廟堂車服禮器，諸生以時習禮其家，余祗迴留之不能去云。天下
君王至于賢人眾矣，當時則榮，沒則已焉。孔子布衣，傳十餘世，學者宗之。自天子王
侯，中國言六藝者折中於夫子，可謂至聖矣。269
As Sima Qian’s description of his travels reveals, Kongzi was much more than a purely textual
presence at this moment in the Western Han. The figure of Kongzi united an entire community
of ritual practitioners whose attachment to the historical Kongzi was presumably no less intense
than Sima Qian’s.
That same attachment seems to have motivated the following episode from the “Zhongni dizi
liezhuan”:
After Kongzi died his disciples missed him dearly. Youruo resembled Kongzi and so Kongzi’s
disciples together set him up as their master and attended to him just as they had attended to
Kongzi. One day the disciples entered to ask, “Previously when the Master was about to
depart he had us carry rain gear, and before long it rained. A disciple asked him, ‘How did you
know that it would rain, Master?’ The Master said, ‘Does not the Shi say, “When the moon is
in the Hyades there will be torrential rains.”’270 Last night wasn’t the moon in the Hyades?’
269. Shiji 47.1947. It should be noted that Huan Tan 桓譚 (43 BCE–28 CE) doubted the authenticity of Taishigong
yue 太史公曰 (The Grand Scribe says) comments like this one, arguing in a fragment preserved in the Quan Hou
Han wen (15/4b; p. 549) that they were authored by Dongfang Shuo 東方朔, an official active during Emperor Wu’s
reign.
270.Note that, once again, Kongzi’s seemingly preternatural knowledge derives from his decidedly mundane
familiarity with texts.
126
Another day, the moon was in the Hyades but it didn’t rain. Shang Que was old and childless
and his mother arranged another wife for him. Kongzi sent him to Qi but his mother begged
him not to. Kongzi said, ‘Do not worry! Shang Que will have five sons after he is forty.’ Sure
enough Kongzi turned out to be right. We ask you, how did the Master know these things?”
Youruo was silent and could not answer. The disciples all arose and said, “Master You should
retire. This is not your seat!”
孔子既沒，弟子思慕，有若狀似孔子，弟子相與共立為師，師之如夫子時也。他日，弟
子進問曰：昔夫子當行，使弟子持雨具，已而果雨。弟子問曰：夫子何以知之？夫子
曰：詩不云乎？月離于畢，俾滂沱矣。昨暮月不宿畢乎？他日，月宿畢，竟不雨。商瞿
年長無子，其母為取室。孔子使之齊，瞿母請之。孔子曰：無憂，瞿年四十後當有五丈
夫子。已而果然。問夫子何以知此？有若默然無以應。弟子起曰：有子避之，此非子之
座也！271
The episode’s appeal for the intended audience of Sima Qian’s Kongzi biography, who, like
Kongzi’s disciples, may have “dearly missed” Kongzi, seems clear. Yet it is striking that
Kongzi’s disciples fail miserably in their attempt to set up Youruo 有若, whose name translates
as “Having a Likeness,” in Kongzi’s place. Like Sima Qian, who “could not reach [Kongzi] although his heart went toward him” (雖不能至，然心鄉往之), Kongzi’s disciples could not
“reach” (zhi 至) him despite presumably having first-hand knowledge of his words and deeds. In
a shorter version of the story at Mengzi 3A/4, Zengzi scolds the other disciples for failing to understand that Kongzi’s “brilliance could not be outdone” (皜皜乎不可尚已). The greater
Kongzi’s perceived virtue, the greater his unattainability.272 By definition, Kongzi had to be
much more than the sum of his sayings, a figure who transcended the limitations of ordinary
human comprehension.
Other Han authors also struggled with their inability to adequately express Kongzi’s magnificence. A dialogue from Hanshi waizhuan book eight and a number of parallel dialogues in
Shuiyuan book eleven, “Shan shui 善說” (“Good at persuading”), feature interlocutors asking
271.Shiji 67.2216.
272.SBCK 5/14a; KZJY pp. 511–12.
127
Zigong to explain “what [Kongzi] was like” (he ru 何如) and Zigong answering that he “does
not know” (bu zhi 不知).273 In the Hanshi waizhuan version, Zigong then compares describing
Kongzi to ascertaining the height of heaven, the breadth of the earth, or the depth of the oceans.
Zigong’s reputation as the most rhetorically gifted Kongzi disciple (e.g., in Lunyu 11/3) allows
us to Zigong’s failure not simply as a personal shortcoming but as a comment on the inadequacy
of human language to capture Kongzi’s greatness. In a similar vein, Lunyu 17/19 (p. 104) has
Kongzi lamenting the need to rely on “sayings” (yan 言) to communicate his ideas. At the risk of
imputing a modern obsession to ancient authors, I would suggest that such passages reflect a certain pessimism about the prospects of understanding Kongzi from the sayings and stories associated with him.
Nevertheless, early authors did venture descriptions of Kongzi, and I suspect that Kongzi
quotation practice significantly shaped early Kongzi encomia, as in the following passage from
Hanshi waizhuan book five:
Kongzi had the heart of a sage. He moved about in the realm of the Way and of virtue and he
wandered in the province of the formless. He relied on heavenly principles, observed human
particulars, understood the beginnings and ends of things, and comprehended success and
failure. Truly he made humaneness and propriety flourish and suppressed opportunism and
profit-seeking; in so doing he maintained and nourished his [sage-heart]. In his time the House
of Zhou was failing and the Kingly Way had been broken off. The feudal lords governed by
force, the strong robbed the weak, the rich oppressed the poor, the people had no peace, and
no one made rules for them. Ritual and propriety being destroyed, human relationships were
not regulated. Whereupon Kongzi “From the west to the east / from the south to the north /
crawled on his knees to help them.”
孔子抱聖人之心，彷徨乎道德之域，逍遙乎無形之鄉，倚天理，觀人情，明終始，知得
失。故興仁義，厭勢利，以持養之。於時周室微，王道絕，諸侯力政，強劫弱，眾暴
寡，百姓靡安，莫之紀綱，禮儀廢壞，人倫不理。於是孔子自東自西，自南自北，匍匐
救之。274
273.See Hanshi waizhuan 8 (SBCK 8/9a–b) and Shuiyuan 11 (SBCK 19a–20a).
274. Hanshi waizhuan 5 (SBCK 5/1b; KZJY 13/1; p. 274). Translation adapted from Hightower 1952, p. 160 (5/1).
For a similar passage that stresses Kongzi’s authorship of the Chunqiu, see Chunqiu fanlu 17, “Yu xu 俞序” (SBCK
6/5a).
128
Given that the Hanshi waizhuan is likely an early Western Han text at the earliest,275 one obviously cannot take this passage as a first-person account of the historical Kongzi. But let us suppose that the Hanshi waizhuan represents a standard view of Kongzi in the Western Han period—what was this Kongzi based on? Perhaps what the Hanshi waizhuan described was, at least
in part, the textual phenomenon called “Kongzi,” i.e., the Kongzi of Kongzi yue. In other words,
the encomium may be a description of the figure who must have existed in order to legitimize
Kongzi quotation practice.
For instance, the claim that Kongzi “relied on heavenly patterns” (yi tianli 依天理) may be
another way of saying that Kongzi sayings articulated principles of universal relevance. “Observed human particulars” (guan renqing 觀人情) is a useful motto for Kongzi comments on
people, just as the phrase “understanding the ends and beginnings of things, and comprehending
success and failure” (ming zhongshi, zhi deshi 明終始，知得失) might describe Kongzi comments on early anecdotal literature. The phrase “Making humaneness and propriety flourish and
suppressing opportunism and materialism” captures the moral concerns of so many Kongzi sayings. The section beginning “the Kingly Way had been broken off” is a clichéd account of Spring
and Autumn or Warring States history, the socio-political background for the anecdotes that so
many Kongzi comments were attached to. And the Shi citation which expresses Kongzi’s concern for the people might reflect the pervasiveness of Kongzi sayings in early texts. Kongzi did
not belong to any one region or group because Kongzi quotation practice was an inter-state phenomenon.276 Finally, the Hanshi waizhuan ties this praise together with a bit of psychologizing:
275.For the dating of the text, see Loewe (ed.) 1993, pp. 125–128.
276. Compare the Han Feizi book 49 (SBCK 19/3a) description of Kongzi as “a sage of the world” (tianxia
shengren 天下聖人); the Lüshi chunqiu book 14/7 description of Kongzi as having “traveled throughout the world”
(孔子周流海內); Kongzi’s self-description in the Liji’s “Tan gong” (SBCK 3/2b; KZJY p. 518) as “a man of the
north, south, east, and west” (東西南北之人); the Shuiyuan’s statement (SBCK 14/7b) that Kongzi “had no fixed
129
“Kongzi had the heart of a sage” (孔子抱聖人之心).
In a roundabout way we have hit upon an insight associated with Michel Foucault in his seminal essay “What is an Author?”:
[The author] does not develop spontaneously as the attribution of a discourse to an individual.
It is, rather, the result of a complex operation which constructs a certain rational being that we
call “author.” Critics doubtless try to give this intelligible being a realistic status, by
discerning, in the individual, a “deep” motive, a “creative” power, or a “design,” the milieu in
which writing originates. Nevertheless, these aspects of an individual which we designate as
making him an author are only a projection, in more or less psychologizing terms, of the
operations that we force texts to undergo, the connections that we make, the traits that we
establish as pertinent, the continuities that we recognize, or the exclusions that we practice. All
these operations vary according to periods and types of discourse.277
Translated into early Chinese terms, statements about Kongzi were, at least in part, a by-product
of statements attributed to Kongzi. “Kongzigraphy”—my term for the representation of
Kongzi—was contingent on the practice of Kongzi quotation. Occasionally, an author even drew
attention to this contingency with his handling of a Kongzi saying, as in the following Huainanzi
passage:
When Zilu saved a drowning man and received an ox as thanks, Kongzi said, “The people of
Lu will always love to to save others from calamity.” When Zigong ransomed someone but
did not accept payment from the state treasury, Kongzi said, “The state of Lu will not ransom
anyone again.” Zilu received something and encouraged virtue, whereas Zigong declined
something and put a stop to goodness. Kongzi’s perspicacity was such that he used the small
to understand the great and the near to understand the far, and was thereby penetrating in his
assessments.
子路撜溺而受牛謝，孔子曰：“魯國必好救人於患。”子贛贖人，而不受金於府，孔子
曰：“魯國不復贖人矣。”子路受而勸德，子贛讓而止善。孔子之明，以小知大，以近知
遠，通於論者也。278
dwelling” (無定處); and Huan Tan’s (Quan Hou Han wen 14/5a) rhetorical question in the Xinlun: “In the past how
could Zhongni have only been the Kongzi of the state of Lu? He was also the sage of Qi [in the east] and Chu [in the
south]” (昔仲尼豈獨是魯孔子？亦齊、楚聖人也。). But note the Zhuangzi’s (SBCK 5/43b; KZJY 17.1/pp. 384–
86) typically contrarian assertion that Kongzi was simply “an accomplished man from the north” (北方之賢者也).
277. Foucault 1979, p. 150. David Schaberg (2005, p. 15) reaches a similar conclusion about such judgments: “From
one point of view, writings of this sort are loving records of an intellectual ancestor’s views. From another, however,
they amount to personifications of the rhetorical function described above. A judgment that enjoyed some
currency—that was perhaps not quite a commonplace, but close to it—was matched with the figure of an important
personage, and thenceforth personage and saying enjoyed a sort of synergy.”
278. Huainanzi 11 (SBCK 11/2a; KZJY 10/47; p. 229).
130
Kongzi was “perspicacious” and “penetrating in his assessments” because his two sayings were
“perspicacious” and “penetrating.” The sayings themselves were the clearest proof of Kongzi’s
acuity.
At the same time, to claim that representations of Kongzi were completely contingent on
Kongzi quotation practice goes too far. In determining the precise relationship between Kongzi
sayings and the figure of Kongzi we face a chicken-and-egg conundrum. Before the first author
invoked a Kongzi yue saying he presumably would have had some familiarity with a “Kongzi” as
well as an expectation that his audience would recognize this figure. But from that point forward,
the representation of Kongzi must have evolved in a dynamic relationship with Kongzi sayings.
Innovations in Kongzi attributions expanded the imagination of Kongzi, which then produced
more Kongzi attributions, which then expanded that imagination even further.
Another way that Kongzi sayings influenced attitudes about Kongzi was by challenging those
attitudes.279 As we have seen, authors whose ideological commitments required a certain image
of Kongzi sometimes had to refute Kongzi quotations that challenged those commitments. In the
course of those refutations, they developed strategies for defending and promoting their own versions of Kongzi, who came to stand above and beyond the words attributed to him. This phenomenon corroborates another of Foucault’s insights:
The author is the principle of thrift in the proliferation of meaning...[T]he author is not an
indefinite source of significations which fill a work; the author does not precede the works, he
is a certain functional principle by which, in our culture, one limits, excludes, and chooses; in
short, by which one impedes the free circulation, the free manipulation, the free composition,
decomposition, and recomposition of fiction. In fact, if we are accustomed to presenting the
author as a genius, as a perpetual surging of invention, it is because, in reality, we make him
function in exactly the opposite fashion. One can say that the author is an ideological product,
since we represent him as the opposite of his historically real function...The author is therefore
279.Nyaln & Wilson 2010, p. 29: “Without his critics, Confucius would be nothing. Like a portrait bust sculpted by
subtraction, the face of Kongzi emerging in the Han and pre-Han sources is defined by swift strikes aimed at the
Master by his many detractors.”
131
the ideological figure by which one marks the manner in which we fear the proliferation of
meaning.280
While the diversity of Kongzi sayings in the early period testifies to the vibrancy of the genre,
that diversity was also a threat to those like the Mengzi author who sought to promote a purer,
more predictable Kongzi. In the next chapter I chart the rise of the Lunyu Kongzi and the attendant disciplining of Kongzi quotation practice.
280.Foucault 1979, p. 159.
132
CHAPTER TWO:
A Brief History of Confucius Quotation Practice, 220 CE–4th century BCE
A crucial element missing from my survey of Kongzi sayings in chapter one is the historical perspective. Although useful for illustrating the breadth and diversity of early Kongzi sayings, the
modern compiler’s luxury of treating individual Kongzi sayings as members of a single, synchronic corpus also distorts the picture in various ways. Not every kind of Kongzi saying was
available for quotation in every period or every context, either because the evolving image of
Kongzi constrained Kongzi quotation practice in new ways or because of the shifting fortunes of
the texts which served as sources of Kongzi sayings. How did Kongzi quotation practice change
over the course of the early period? What triggered those changes? And what role did specific
texts play in these developments?
Kongzi quotation practice in the last two millennia has been dominated by the Lunyu and to a
lesser extent by Kongzi material contained in the Five (or Six or Seven or Nine or Thirteen)
Classics.281 When one looks back at this tradition of Kongzi quotation and the countless authors
who peppered their texts with Lunyu quotations, references, and allusions, be they emperors or
statesmen, commentators or philosophers, poets or proselytizers, it can seem as though the Lunyu
is as old as the Chinese tradition itself.
But the Lunyu’s authority was not a historical constant. To demonstrate this point, in this
chapter I survey early quotation practice beginning from the end of the early period and moving
backwards through the Eastern Han (25–220), Xin 新 (9–23), Western Han 西漢 (205 BCE–9
281.The term “Seven Classics” (qi jing 七經), i.e., the Six Classics plus the Lunyu, first appears in the Eastern Han
period. For references, see p. 156 below.
133
CE), and Qin (221–206 BCE) empires before concluding in the Warring States period (5th century–221 BCE). I argue that Kongzi quotation practice can be divided into two distinct phases: a
post-Lunyu phase in which authors regularly invoked the Lunyu when quoting Kongzi, and a preLunyu phase in which Kongzi quotations exhibit little to no overlap with Lunyu sayings. The
more familiar textual milieu from the perspective of the later tradition is the Eastern Han, Xin,
and late Western Han, a time when the authority of both Kongzi and of the Lunyu was well-established. But as our survey moves into the early Western Han, Qin, and Warring States periods
the familiar will give way to the unfamiliar as the Lunyu’s influence becomes difficult if not impossible to detect, even when Kongzi’s authority as a source of masterful sayings remains
constant.
Kongzi quotation practice in the early period: The Big Picture
The graphs on pages 139–141 summarize the results of my survey of early Kongzi sayings.
Graph 1 lists the most significant sources of Kongzi quotations from the Eastern Han (25–220
CE) period, graph 2 those from the Western Han (205 BCE–9 CE), and graph 3 from the preHan era.282 Blue (or dark grey if not viewed in color) bars represent the total number of Kongzi
attributions in a given text, green (or light grey) bars represent Kongzi quotations which have
close parallels in the Lunyu and thus might be considered Lunyu quotations. Numbers within the
diamonds indicate Kongzi sayings with Lunyu parallels as a percentage of the total number of
Kongzi quotations in a given text. Graded yellow (or graded grey) bars indicate the number of
282.By “significant sources” I mean texts that quote Kongzi more than once or twice.
134
explicit Lunyu attributions, i.e., quotations marked by Lunyu yue 論語曰 (“the Lunyu says”) or
some variation thereof.
I have made every effort to arrange texts in chronological order from bottom to top. However, the uncertain dating and heterogeneous nature of so many early texts makes this difficult, if
not possible.283 For texts whose chronology is relatively certain I have included tentative dates to
serve as points of reference.284 However, the near certainty that many of these texts evolved into
their present form over the course of the early period and beyond makes all such dates extremely
provisional. For a few large compilations, e.g., the Liji and Shiji, I have listed the data for specific chapters as well as for the entire text.
My method for counting Kongzi attributions and for identifying attributions with Lunyu parallels will become clearer in my discussion of specific examples below, but a few points should
be noted at the outset. First, my data for Kongzi attributions and Lunyu parallels include explicit
attributions (i.e., sayings prefaced by Kongzi yue 孔子曰 and its variants) as well as possible implicit attributions, e.g., when Mengzi 7B/37 describes Kongzi with words attributed to Kongzi in
the Lunyu: “Mengzi said, ‘Kongzi could not find men of the middle Way to associate with...’”
(孟子曰：孔子不得中道而與之...).285 Even without the Kongzi yue quotation marker, here the
explicit association with Kongzi at least marks it as a possible Lunyu quotation. Unattributed
Lunyu parallels that do not at least mention Kongzi in some way have not been included in my
283.Although it is in need of revision, Michael Loewe’s (ed.) Early Chinese Texts: A Bibliographical Guide is still
the best introduction to the chronology of early received texts.
284.My (rather cautious) approach here might be contrasted with that of Paul Fischer in his 2009 article
“Intertextuality in Early Chinese Masters-Texts: Shared Narratives in Shi Zi.” Despite his thoughtfulness on the
question of parallels, Fischer seems to take for granted that early masters-texts can be dated according to the lifespan
of their eponymous authors.
285.Mengzi 7B/37 (SBCK 14/16a–b; KZJY pp. 516–17). I discuss Mengzi 7B/37 in greater detail on p. 204.
135
data set. I address such parallels at the conclusion of the chapter.
As a general rule, the criteria for determining whether or not two texts are “parallel” will
vary according to the interests of the surveyor.286 Since I am primarily concerned with determining the extent of the Lunyu’s influence and circulation in the early period, the phrase “Lunyu
parallel” will mean “a piece of text which is similar enough to the Lunyu that one might plausibly interpret it as a quotation of the Lunyu.” Here the word “might” is crucial: a Lunyu parallel is
only a candidate Lunyu quotation.287 As I will argue below, there were Lunyu parallels long before there were actual Lunyu quotations.
In my terminology, a “Lunyu parallel” is both verbally and syntactically similar to a Lunyu
saying.288 It will share at least three words with a Lunyu entry (although graphic and semantic
variants are permissible) and those words will be related to one another in a similar way.289 Similar themes or ideas will not count as “parallels” unless they are expressed in similar language.
For instance, book 22 of the Xunzi, “Zheng ming 正名” (“Rectifying names”), contains four instances of the two-word phrase zheng ming 正名, which also appears at Lunyu 13/3 (“I would invariably rectify names!” [bi ye zheng ming hu 必也正命乎]). But since the specific context in the
286.Fischer 2009, p. 4: “Judging which passages are ‘similar enough’ to warrant being counted as ‘parallel’ is a
subjective matter and will always be problematic.”
287.Schultz 1999, p. 142: “[N]ot every parallel is a quotation or involves some type of literary dependence.
Formulae and stereotypical phrases often appear to be similar to quotations, but simply reflect standardized
expressions for describing characteristics, gestures, common actions, as well as repeated natural phenomena.
Proverbs are also problematic since they combine striking formulation with a complete, though often generally
applicable, thought. Proverbs may be considered to be a type of quotation, since similar introductory formulae are
sometimes used. Yet the ideas of origin or authorship and of context, which are inherent elements of true quotation,
are lacking.”
288.Here I follow Schultz (1999, p. 222).
289.Compare Schultz (1999, p. 19, 217) on “verbal parallels.” For Schultz as well, verbal parallelism does not imply
“verbal dependence,” i.e., quotation proper. He prefers to restrict quotation to instances “in which an exegetical
purpose in reusing earlier material can be demonstrated or where an understanding of the earlier text and context is
helpful, if not essential, for a proper interpretation of the new text” (p. 221ff). Christiane Haupt (2006, p. 22) in her
study of Lunyu parallels uses a four-character standard.
136
Xunzi looks nothing like the Lunyu version, I do not count the phrase zheng ming 正名 as a
“parallel.”290
The high degree of variability among parallel Kongzi sayings has forced me to make numerous subjective decisions about whether a given Kongzi quotation truly counts as a Lunyu parallel. In the case of Eastern Han texts, for which the question of the Lunyu’s influence can be answered straightforwardly, it matters relatively little whether one among hundreds of possible
Lunyu quotations counts as a Lunyu parallel. But for the Warring States and early Western Han,
periods for which the existence of the Lunyu is very much in doubt, a great deal hinges on the
analysis of a handful of parallels. Consequently, when identifying possible Lunyu parallels in
Warring States and Western Han texts I have used a much laxer standard than for later texts, with
the result that my data for Lunyu parallels in Graph 3 also include Kongzi attributions which vary
significantly from their Lunyu counterparts.
Consider the parallels to Lunyu 13/18 listed in Appendix 2:A. In the progression from the late
third-century Lüshi chunqiu passage to the Western Han Hanshi waizhuan and Eastern Han Baihu tongyi we see a typical example of Lunyu parallel variability. Even when Warring States-era
parallels can be found for a given Lunyu passage, those parallels tend to have more numerous
and more significant variants than parallels from later periods. Although the Lüshi chunqiu and
Lunyu versions share a similar context (the Upright Self story) and interest (father-son relations
and the question of “honesty” [xin 信] or “uprightness” [zhi 直]), the differences between them
(e.g., using the Duke of She as an interlocutor instead of the king of Chu) and the accompanying
Kongzi sayings are substantial enough to make the Lüshi chunqiu version seem an unlikely Lun-
290.See Fischer 2009 for a list of 24 kinds of intertextuality in early Chinese texts. Since my goal in this chapter is
to date the Lunyu and not to study intertextuality per se, I am not concerned with such fine-grained distinctions.
137
yu quotation. In contrast, save for a slight variation in the order of its Kongzi saying (“sons and
fathers” instead of “fathers and sons”) the Hanshi waizhuan version is a word-for-word quotation
of Lunyu 13/18. The even later Baihu tongyi not only corrects the variant in the Hanshi
waizhuan, it removes any doubt about the origin of the saying by naming its source: “The Lunyu
says” (Lunyu yue 論語曰). In my survey of Kongzi quotations, all three versions count as “Lunyu parallels” even though the Lüshi chunqiu version is probably not a “Lunyu quotation,” i.e., a
conscious borrowing from a Lunyu text.
*
*
138
*
Graph 1: Kongzi sayings in Xin and Eastern Han texts (9 CE–220 CE)
Hou Hanshu 後漢書
78
37
8
7
88%
22
1
1
100%
14
22
Fengsu tongyi 風俗通義
24
32
25
36
5
4
80%
14
43
27
Wang Yi’s 王逸
Chuci 楚辭 commentary
63%
11
13
2
15%
33
Lunheng 論衡, c. 100
72
63%
115
6
Lunheng 論衡 book 28:
“Wen Kong 問孔”
14
13
Hanshu 漢書, c. 92
93%
146
8
20
Baihu tong 白虎通
5
25%
47
5
1
20%
5
13
12
Eastern Han edicts
Xinlun 新論, 26 CE
75%
27
1
Qianfu lun 潛夫論
書
78%
30
Xu Shen’s 許慎 / Gao You’s 高誘
Huainanzi 淮南子 commentary
Yuejue shu 越
41%
9
Zheng Xuan’s 玄 (127–200)
san li 三禮/Shi 詩 commentaries
Shuowen jiezi 說文解字, c. 100
62%
23
Gao You’s 高誘
Lüshi chunqiu 呂氏春秋 commentary
He Xiu’s 何休 (129–182)
Gongyang 公羊 commentary
115
2
Zhonglun 中論
Zhao Qi’s 趙 (d. 201)
Mengzi 孟子 commentary
68%
5
4
92%
80%
139
84% 174
Graph 2: Kongzi sayings in Western Han texts (205 BCE–9 CE)
Shuiyuan 說苑, 17 BCE
Xinxu 新序, c. 24 BCE
22
13
1
Da Dai liji 大戴禮記
Yantie lun 鹽鐵論
Western Han memorials
Hanshi waizhuan 韓詩外傳
Shangshu dazhuan 尚書大傳
Liji 禮記
8
5
77%
43
35
81%
13%
14
27
3
1
112
11%
9%
28
1
3
“Zhongni yanju 仲尼燕居”
3
46
16
8
7%
19%
39
5
1
“Zhongyong 中庸”
“Biao ji 表記”
83%
63%
3
“Tan gong 檀弓”
“Fangji 坊記”
42
35
30
23
3
133
6%
4
22
13%
36%
51
2
4%
Shiji 史記, c. 100 BCE
148
Shiji 史記 books 47 & 67
124
Shiji: all books except 47 & 67
Huainanzi 淮南子, c. 139 BCE
48
24
21
3
9
5
Guliang 穀梁
7
0%
8
13%
1
Xiaojing 孝經
1
Xinshu 新書
1
Mawangdui manuscripts,
c. 168 BCE
6
234
186
50%
56%
15
67%
63%
14%
Xinyu 新語
Gongyang 公羊
181
59%
8
Chunqiu fanlu 春秋繁露
Western Han edicts
13%
24
Total Confucius attributions
Confucius attributions that parallel the Analects
Total Analects attributions
Analects Confucius attributions as a percentage
of total Confucius attributions
7%
17%
0%
140
116
320
Graph 3: Kongzi sayings in pre-Han texts (5th century–205 BCE)
Zhuangzi 莊子
Yanzi chunqiu 晏子春秋
3%
3
7%
1
15
Zhouyi 周易
0%
Lüshi chunqiu 呂氏春秋,
c. 239 BCE
Han Feizi 韓非子
Xunzi 荀子
“Zi yi 緇衣”,
c. 300 BCE
3%
1
8%
2
6
11
Mozi 墨子
25
0%
7%
3
9
2
1
42
8%
6
Mengzi 孟子
38
0%
Guoyu 國語
Shanghai Museum
manuscripts, c. 300 BCE
33
5%
2
Zuozhuan 左傳
34
12%
4
“Min zhi fumu 民之父母”,
c. 300 BCE
30
27%
33
50%
141
78
92
*
*
*
A significant shift in Kongzi quotation practice is evident in the contrast between the Eastern
Han (graph 1) and pre-Han eras (graph 3). Although a wealth of texts in both periods quote
Kongzi and/or the anonymous Master, Warring States texts contain zero instances in which an
author explicitly attributes something to the Lunyu. Even more remarkably, Warring States texts
include a much lower percentage of Kongzi sayings with Lunyu parallels compared to Eastern
Han texts. By my (very rough) count, roughly 70% of Eastern Han Kongzi attributions parallel
the Lunyu as opposed to 9% in the pre-Han era. Keep in mind that the 9% figure is significantly
inflated because it also includes parallels with substantial variants like the Lunyu 13/18 parallel
in the Lüshi chunqiu discussed above.
If the Warring States and Eastern Han periods represent two extremes of Kongzi quotation
practice, the Western Han period falls somewhere in between. Some texts, e.g., the Shiji, look
more like Eastern Han texts insofar as they include a relatively high percentage of Lunyu parallels. On the other hand, texts like the Huainanzi and Liji do not appear to privilege Lunyu Kongzi
sayings over non-Lunyu sayings. The Shiji/Huainanzi comparison is particularly striking. Both
texts are very large compendia (just under 600,000 characters for the Shiji and over 130,000
characters for the Huainanzi), both were conceived as comprehensive summa, and both were
compiled (at least in part) during the reign of Emperor Wu (141–87 BCE). Nonetheless, these
two texts use Lunyu Kongzi sayings to very different degrees very differently to Lunyu Kongzi
sayings, with 63% of the Shiji’s Kongzi sayings having parallels in the Lunyu compared to 14%
of the Huainanzi’s. Thus it would seem that the few decades between the submission of the
Huainanzi in 139 BCE at the beginning of Emperor Wu’s reign and the compilation of the Shiji
towards the end of Emperor Wu’s reign marked a turning point in the influence of Lunyu Kongzi
142
sayings. Early Western Han texts (with the possible exception of Lu Jia’s 陸賈 [d. c. 140 BCE]
Xinyu 新語)291 do not appear to privilege Lunyu Kongzi sayings, whereas late Western Han texts
make frequent use of Lunyu Kongzi sayings and also quote the Lunyu explicitly.
But these numbers can only take us so far. As we will see below, post-Lunyu authors did not
just quote the Lunyu more often than pre-Lunyu authors, they also treated it as if it was an authoritative and recognizable text. The Lunyu’s absence in the pre-Han era is felt just as keenly in
the way that pre-Han authors handled Kongzi sayings compared to post-Lunyu authors.
The Eastern Han (25–220 CE)
• Edicts
Let us begin with the event that formally ended the Eastern Han period, Cao Pi 曹丕’s (187–226)
accession in 220 CE as Emperor Wen of the Wei dynasty 魏文帝 (r. 220–226). Even though the
Liu 劉 clan had long since lost de facto control of the empire, declaring an end to the Han
dynasty was no trivial matter. Cao Pi’s father, the warlord Cao Cao 曹操 (155–220), had begun
laying the groundwork for this transition in the years leading up to his death on March 15, 220.
Having already seized the nominal Han Emperor Xian 漢獻帝 (r. 189–220) in 196, in 216 he declared himself King of Wei 魏王 besides assuming a number of other imperial prerogatives.292
Cao Pi’s official accession on December 11th, 220, was preceded by a month-long rhetorical ex-
291.The dating of this particular text is extremely problematic. See Loewe (ed.) 1993, pp. 171–177, for critiques of
its authenticity.
292.Goodman 1998, p. 56, has a useful summary of these steps.
143
change in which Emperor Xian and Cao Pi’s officials cited various omens and historical precedents to urge him to ascend the throne as emperor,293 and Cao Pi, in turn, coyly refused, professing his inferior virtue and his reverence for heaven’s mandate (tian ming 天命).
The Cao Pi accession materials are a veritable sourcebook of dynastic legitimacy detailing
the precedents, political theories, and textual authorities that truly mattered to Cao Pi and his
multiple constituencies. Not surprisingly, one of the more frequently invoked authorities in these
materials was Kongzi.294 On November 23rd, 220, a group of officials headed by Investigator of
the Army and Palace Assistant Sima Yi 司馬懿 (179–251), who in 249 would wrest control of
the Wei dynasty from the Cao clan, submitted a memorial which included this quotation of Lunyu 9/9:
We subjects have heard that when the era of Yao went into decline the mandate of heaven was
then to be found in Shun. When Shun’s era declined the mandate was to be found in Yu. This
being the case, the numinous power of heaven and earth, calendrical cycles, and omens
concerning the dismissal and appointment of [rulers] depend solely on virtue. Thus Kongzi
said, “The phoenix does not come, the river gives forth no chart. It is all over with me!”
臣等聞有唐世衰，天命在虞，虞氏世衰，天命在夏；然則天地之靈，曆數之運，去就之
符，惟德所在。故孔子曰：鳳鳥不至，河不出圖，吾已矣夫！295
Even by this time, the memorial’s core argument—that rulers like Cao Pi should respect heaven’s mandate—was one of the oldest clichés in the Chinese tradition. But its invocation of
Kongzi was rather artful. According to one early commentator, Lunyu 9/9 expressed Kongzi’s
“grief that he could not see” (shang bu de jian 傷不得見) a sage-ruler in his lifetime whose
virtue was great enough to trigger the auspicious omens of the phoenix and river chart.296 Sima
293.According to the Sanguo zhi (2.62), the event that precipitated this exchange was Emperor Xian’s 漢獻帝
determination on November 19th that “the hopes of the masses depend on Wei” (zhong wang zai Wei 眾望在魏).
294.Cao Pi also established regular sacrifices to Kongzi in an edict dated to 221. See Sanguo zhi 2.77–78.
295.Sanguo zhi 2.66. Translation adapted from Goodman 1998, p. 112. In another memorial begging Cao Pi to take
the throne, his Three Excellencies (san gong 三公) explicitly cited Lunyu 16/8 to remind him that “the Lunyu says
that ‘the junzi fears Heaven’s mandate’” (論語云：君子畏天命). See Sanguo zhu 2.73.
296.This is the Kong yue 孔曰 layer of the Lunyu jijie 論語集解 (Collected Explications of the Lunyu). See Lunyu
144
Yi used Lunyu 9/9 to contrast Kongzi’s situation with his own: unlike Kongzi, Cao Pi’s subjects
could rejoice that they lived under a ruler as virtuous as Cao Pi—provided, of course, that Cao Pi
would actually acknowledge his virtue and claim the mandate.
In his reply to Sima Yi et al., Cao Pi quoted Lunyu 4/14 to rebut Lunyu 9/9:
What this age lacks is the Way and propriety, but it has negligence and foolishness in
abundance. The average person is predisposed to belittle what he lacks and to value what he
has in abundance. Thus it is said, “One does not worry about lacking an official position; one
worries about having the means to establish oneself.” Since I have little virtue, I hope to avoid
the average person’s [fault of] valuing [what he has in abundance].
世之所不足者道義也，所有餘者苟妄也；常人之性，賤所不足，貴所有餘，故曰：不患
無位，患所以立。孤雖寡德，庶自免于常人之貴。297
Against Sima Yi’s invocation of Kongzi as an authority on the virtue of rulers Cao Pi pointed to
Kongzi as a model for prioritizing the cultivation of personal virtue over official service. Kongzi
may have urged the rulers of his day to heed the mandate of heaven, Cao Pi argued, but he also
urged individuals to “establish” (li 立) themselves before all else. Cao Pi appealed to this Kongzi
again when he concluded the same memorial with a quotation of Lunyu 9/26: “Thus it is said,
‘One can rob the Three Armies of their general but one cannot rob a commoner of his will.’ How
can I be robbed of this will of mine?” (故曰：三軍可奪帥，匹夫不可奪志。吾之斯志，豈可
奪哉？).298 The Lunyu was apparently a natural choice to communicate Cao Pi’s modesty and
steadfastness in the face of his officials’ flattery, a pose which ultimately enhanced his image as
a uniquely virtuous ruler.299
Cao Pi was breaking with the Han dynastic tradition at this moment, but in his use of the
jishi p. 589.
297.Sanguo zhi 2.66–67.
298.Sanguo zhi 2.67.
299.While the Lunyu dominated Kongzi quotation practice at the imperial level, it was not the only text that
emperors appealed to when quoting Kongzi. Cao Pi in another document referenced a Kongzi anecdote from the
Zhuangzi in which Kongzi is said to have taken a certain Wang Tai 王駘 as his teacher.) See Sanguo zhi 2.69.
145
Lunyu he certainly did not break with Han rhetorical tradition of invoking Kongzi and the Lunyu
on the imperial stage. Cao Pi’s father Cao Cao and several Eastern Han rulers before him were
accustomed to quoting Kongzi and the Lunyu in their official proclamations, just as their officials
quoted Kongzi and the Lunyu in their memorials. According to one early source, quoting Kongzi
was not simply a matter of rhetorical ornamentation. A 31 CE memorial submitted by Zhu Fu 朱
浮 (d. 57), one of Emperor Guangwu’s 光武帝 (r. 25–57) closest allies in the wars which preceded the establishment of the Eastern Han, urged the emperor to commit himself to the imperial
academy (taixue 太學) and the selection of erudites (boshi 博士) “so as to ensure that the sayings
of the sage Kongzi are transmitted without end” (使孔聖之言傳而不絕。).300 Judging from extant Eastern Han edicts, Guangwu and later emperors took this advice to heart (see Appendix
2:B). All told, of the eighteen extant Eastern Han edicts which quote or explicitly reference
Kongzi, sixteen draw their Kongzi quotations from the Lunyu.301 Eastern Han emperors’ near exclusive use of the Lunyu when quoting Kongzi is one of the surest proofs of its authority in the
period. Not even the Kongzi sayings preserved in the Five Classics could compete with Lunyu
Kongzi sayings on an imperial stage.302
This edict issued by Emperor Zhang 章帝 (r. 57–88 BCE) in 80 CE quoted Lunyu 13/3, the
very same Kongzi saying invoked by Emperor Guangwu in 26 CE (p. 101):
<A>Kongzi said, “When punishments are wrongly applied the people have nowhere to place
their hands and feet.” <B>Nowadays many officials are immoral. They arrogate authority and
act only for their own gratification or out of anger. When handling cases some disregard
300.Hou Hanshu 33.1144.
301.A handful of other sayings incorporated Lunyu sayings without attribution. See the edicts of 70 CE (Hou
Hanshu 2.117) and 79 CE (Hou Hanshu 10A.414).
302.The Hou Hanshu records that three Eastern Han emperors, Emperors Ming 明帝 (r. 58–75), Zhang 章帝 (r. 75–
88) in 85, and An 安帝 (r. 106–125), personally sacrificed to Kongzi. See Hou Hanshu 2.118, 3.149–150, and 5.238.
In 178 CE, paintings of Kongzi and his seventy-two disciples were placed in the Hongdu Palace 鴻都門 (Hou
Hanshu 50b.1998). On the development of the cult of Kongzi in the Han, see also Shryock 1966, Huang Jinxing
1994, pp. 185–246, and Nylan & Wilson, pp 72–73.
146
suspects’ guilt or innocence, and they oppress the innocent to the point that those driven to
suicide in a single year outnumber those who receive a verdict. This is certainly not what is
meant by “being a father and mother to others.”303 <C>Let the appropriate authorities debate
the impeachment of such officials.
孔子曰：刑罰不中，則人無所措手足。今吏多不良，擅行喜怒，或案不以罪，迫脅無
辜，致令自殺者，一歲且多於斷獄，甚非為人父母之意也。有司其議糾舉之。304
The edicts preserved in the Han histories typically consist of three sections: (A) a statement of
principles, (B) a critique of contemporary circumstances (usually marked with jin 今 [nowadays]) in light of those principles, and (C) the final command (marked with the modal particle qi
其).305 Kongzi sayings were not specific or relevant enough to describe contemporary affairs, and
only the emperor’s own words had the requisite authority to serve as commands. But Kongzi
sayings were tailor-made sources for the general principles which ostensibly motivated imperial
policy-making. Put simply, emperors’ Kongzi quotations were proof texts, quotations which corroborated the thinking behind the policies. This is precisely how the quotation of Lunyu 13/3
functions in Emperor Zhang’s edict above. Juxtaposed with his description of contemporary judicial practices, the Kongzi saying underlined Emperor Zhang’s dissatisfaction with the status quo
and cast his policy announcement as a sincere effort to bring contemporary governance in line
with classical ideals.306 When uttered by emperors, quotations like this one must have also rein-
303.This is a reference to the Mao Shi #251, “Jiong zhuo 泂酌” (“At the Wayside Pool”).
304.Hou Hanshu 3.140. Loewe 2004, pp. 522–46, is an excellent overview of Han imperial edicts. One must always
keep in mind when dealing with Han edicts that “with some very few exceptions, the texts of decrees that are
included in the Shiji and the Hanshu are abbreviated versions of original documents” (p. 528). Moreover, “not all the
decrees that an emperor or his advisors issued were included in the histories” (p. 548).
305.However, see Sanft 2007–2008 for a study and translation of a late Western Han edict discovered on the wall of
an outpost at Xuanquan 懸泉, Dunhuang 敦煌 commandery. If this document, “the sole example of a more or less
complete, Han-era imperial edict” (p. 126), is at all representative of actual edicts from the period, then the received
edicts preserved in the histories must have originally consisted of many more sections.
306.Loewe 2004, p. 522: “Citing as they do from texts that were coming to be revered as sacred, or referring as they
may to the ethical ideals ascribed to the masters, some decrees reveal a growing respect for the authority and lessons
of the past.”
147
forced the Lunyu’s status as a uniquely authoritative source of Kongzi sayings.
• Commentaries
The proliferation of Lunyu commentaries in the Eastern Han is yet another index of the text’s
prestige in the period. The Lunyu jijie 論語集解 (Collected Explications of the Lunyu), a text
thought to have been compiled by He Yan 何晏 (190–249) not long after the fall of the Eastern
Han,307 preserves selections from eight Han-era Lunyu commentaries.308 Less important for our
purposes than the substance of these commentaries is the mere fact that the Lunyu had attracted
so much scholarly interest by the end of the Han. Two of these Lunyu commentators, Ma Rong
馬融 (79–166) and Zheng Xuan 鄭玄 (127–200), were among the most illustrious scholars of
their time.309
Commentaries on other texts are in many ways a more revealing measure of the Lunyu’s influence because they show the extent to which it had come to permeate Eastern Han scholarly
discourse. The table at Appendix 2:C describes the distribution of Kongzi sayings in nine extant
commentaries from the Eastern Han, all of which regularly attribute quotations to both Kongzi
307.See Makeham 1999 for the argument that the Lunyu jijie was not the work of a single compiler.
308.According to the Lunyu jijie preface (SBCK 1/1a–3a; Shisan jing zhushu 11.2454ff), these are: Kong Anguo’s
孔安國 (fl. Emperor Wu’s reign, 141–87 BCE) xunjie 訓解 (glosses and explanations), the zhangju 章句 (chapterand-verse commentary) of Mr. Zhou 周氏 and Bao Xian 包咸 (d. 65 CE), Ma Rong’s (79–166) xunshuo (glosses
and explanations), Zheng Xuan’s 鄭玄 (127–200) zhu 注 (commentary), and the yishuo 義說 (explanations of the
meaning) of Chen Qun 陳群 (d. 236), Wang Su 王肅 (195–256), and Zhousheng Lie 周生烈 (195–256). With the
exception of Kong Anguo, all of these men lived during the late Western Han or later. But the identification of the
Kong yue 孔曰 (Kong said) layer of commentary with Kong Anguo is problematic. The “Yiwen zhi” bibliography
(Hanshu 30.1716–17) preserves no mention of a Kong Anguo bibliography even though it elsewhere credits Kong
Anguo, a supposed descendent of Kongzi, with having “obtained” (de 得) and “submitted” (xian 獻) several
Kongzi-related texts.
309.For brief biographies of Ma Rong and Zheng Xuan, see de Crespigny 2007, pp. 648–49 & 1126–28. For a study
of Zheng Xuan’s Lunyu commentary, see Makeham 1997.
148
and the Lunyu. While 64 (78%) of their 82 Kongzi yue sayings are likely Lunyu quotations, they
also explicitly quote the Lunyu in 94 instances.310
In contrast to Eastern Han emperors, Eastern Han commentators used Kongzi sayings not as
proof texts but as useful points of reference. For commentator and audience alike, the Lunyu was
a reliable touchstone for contextualizing the language and ideas of the texts commented on. In
this respect, it was valued first and foremost as a source of recognizable language and concepts,
not as a source of Kongzi’s teachings. This function also helps to explain the form of commentators’ Lunyu quotations, which are often abbreviated and thus rather opaque for readers unfamiliar
with the source.
Consider the following examples from Wang Yi’s 王逸 (fl. c. 120 CE) Chuci commentary,
the Xu Shen 許慎 (c. 55–c. 149)/Gao You 高誘 (c. 168–212) Huainanzi commentary,311 and
Zhao Qi’s 趙岐 (129–82) Mengzi commentary:
Li sao 離騷
(Encountering Sorrow)
Wang Yi’s commentary
It is the reckless pleasure of the “Cliques” are companions. Lunyu [15/22] says, “Be compancliques.
ionable without being cliquish.” “Reckless” means careless.
惟夫黨人之偷樂兮...
黨，朋也。論語曰：朋而不黨。偷，苟且也。312
310.Zhao Qi’s preface adds four more Kongzi sayings (three of which parallel the Lunyu) and He Xiu’s Gongyang
preface adds one additional Kongzi saying of unknown origin.
311.The Xu Shen/Gao You commentaries were woven together at some point in their transmission history, making
the identification of distinct layers. See Loewe (ed.) 1993, pp. 190–191.
312.SBCK 1/9a.
149
Huainanzi 16,
“Shui shan 說山”
(“Mountain of Persuasions”)
Xu Shen/Gao You commentary
“Inconsistent actions” is like Lunyu [13/22]: “An inconsistent
Thus people hate nothing more person cannot be a medium or doctor.” This is why the text
than inconsistent actions.
says [such actions are] hateful.
故人莫惡於無常行。
無常行猶論語：人而無恆，不可作為巫醫。故曰惡矣。313
Zhao Qi commentary
Mengzi book 5 title,
“Wan Zhang 萬章”
Wan Zhang [in the first episode of Mengzi book 5] asks about
the filiality of Shun just as Yan Yuan asks [Kongzi] about humaneness in [the first episode of] Lunyu [book 12], and because of this [“Wan Zhang”] was used as the chapter title.
萬章問舜孝，猶論語顏淵問仁，因以題其篇也。314
In the first example, Wang Yi references Lunyu 15/22 (“The Master said, ‘The noble man is serious without being quarrelsome, he forms groups without being cliquish” [子曰：君子矜而不
爭，群而不黨。]) in order to pin down the meaning of the word dang 黨 (clique). His point is a
semantic one: dang and peng 朋 (companion) are roughly synonymous except that dang in the Li
sao is the morally problematic or pejorative counterpart of peng, just as it is in the Lunyu. While
the Master of Lunyu 15/22 is in a prescriptive or moralizing mode when he distinguishes these
words, Wang Yi seems to care little about ethics. His point is simply that familiarity with Lunyu
15/22 makes this particular Li sao line more comprehensible.
The use of the word you 猶 (to be like) in the second and third examples perfectly captures
the appeal of Lunyu quotations for Eastern Han commentators, who did not set out to defend the
reasoning behind a given line so much as point out an illuminating convergence. The quotation
of Lunyu 15/22 rendered the Huainanzi line more intelligible, and the reference to the title of
313.Huainanzi 16 (SBCK 16/14a).
314.Mengzi 5a/1 (SBCK 9/1a).
150
Lunyu book 12 helped to contextualize the choice of a chapter title in the Mengzi. The Lunyu’s
appeal for Xu Shen as author of the Shuowen jiezi 說文解字 (Explanations of Simple and Complex Characters) also seems to have had more to do with its recognizability than its substance,
e.g., in this Shuowen jiezi entry on the character fen 份:
Fen 份: the patterning and substance are complementary. [The character form] follows ren 人
and it sounds like fen 分. The Lunyu says, “The patterning and substance are fenfen.”
份：文質備也。從人分聲。論語曰：文質份份。315
To explain snippets of language like this one Xu Shen needed a well known text, not an authority. These are Lunyu quotations first and Kongzi sayings second.
For Zhao Qi, however, you 猶 seems to have indicated something more than superficial similarity. In the preface to his Mengzi commentary we find this passage:
After Kongzi returned to Lu from Wei, true music was rectified and the Elegentiae and
Eulogies [divisions of the Shi] each obtained their proper place. Then Kongzi edited the Shi
and Shu, appended [commentaries] to the Yi, and composed the Chunqiu. When Mengzi
returned from Qi and Liang he transmitted the way of Yao and Shun and composed writings
thereby. This was an instance of a man of great worth composing in imitation of a sage. The
seventy disciples gathered the Master’s sayings together and made the Lunyu. The Lunyu is
the linch-pin of the Five Classics and the mouthpiece of the Six Arts. When Mengzi wrote his
text he modeled it on and imitated the Lunyu. When Duke Ling of Wei asked about military
formations, Kongzi answered with ritual vessels; when King Hui of Liang asked about
benefiting his state, Mengzi answered with humaneness and propriety. When Huan Kui of
Song wanted to harm him, Kongzi said, “heaven birthed the virtue in me”; when Zang Cang
slandered him, Mengzi said, “How could the son of the Zang clan have caused me not to meet
with [an opportunity to serve as an advisor to the Duke of Lu]?” There are many examples like
these in which Kongzi’s and Mengzi’s meaning and intentions are in agreement.
孔子自衛反魯，然後樂正，雅、頌各得其所，乃刪詩定書，繫周易，作春秋。孟子退自
齊、梁，述堯、舜之道而著作焉，此大賢擬聖而作者也。七十子之疇，會集夫子所言，
以為論語。論語者，五經之錧鎋，六藝之喉衿也。孟子之書，則而象之。衛靈公問陳於
孔子，孔子答以俎豆；梁惠王問利國，孟子對以仁義。宋桓魁欲害孔子，孔子稱天生德
於予；魯臧倉毀鬲孟子，孟子曰：臧氏之子焉能使予不遇哉！旨意合同，若此者眾。316
The Lunyu for Zhao Qi was much more than a useful point of reference. Although references to
315.SBCK 15/3b. This Lunyu quotation happens to be a variant of Lunyu 6/18 (文質彬彬，然後君子。).
Reduplicative binomes are highly unstable in early texts so it is not surprising to find such variations even in Eastern
Han texts. On this point see Knechtges 1987, pp. 1–13, and Kern 2007, p. 784.
316.See the front matter to the SBCK edition, pp. 3a–3b.
151
and quotations from a Mengzi text can be found in some texts from the Han era,317 the Mengzi did
not enjoy anything like the Lunyu’s authority or visibility in the period. By noting the similarities
between the Lunyu and his version of the Mengzi, and between Kongzi and Mengzi themselves,
Zhao Qi apparently hoped that his text would ride the Lunyu’s coattails.318
Although not a commentary per se, another text which confirms the Lunyu’s authority in
scholarly circles is the Baihu tongyi 白虎通義 (General Meaning of the White Tiger Hall [Conference]), ostensibly a digest of a conference convened in 79 CE by Emperor Zhang in order to
“discuss and debate the points of agreement and disagreement in the Five Classics” (講議五經同
異).319 One of the most meticulously sourced texts from the early period, the Baihu tongyi explicitly quotes the Lunyu in 47 instances and includes twenty more unsourced Kongzi attributions,
five of which originated in the Lunyu and eleven in the Liji. Another twenty sourced quotations
(including eight Xiaojing yue 孝經曰 and eight Liji yue 禮記曰) cite sayings which are attributed
to Kongzi in the source texts themselves, as in the following Xiaojing quotation:
<Baihu tongyi>The Xiaojing says, “The former kings saw that their teachings could transform
the people.”
孝經曰：先王見教之可以化民。320
<Xiaojing>The Master said, “...The former kings saw that their teachings could transform the
people.”
子曰：...先王見教之可以化民也。321
317.For a more or less complete list of these quotations, see Citations from the Mengzi to Be Found in Pre-Han and
Han Texts (2007).
318.Zhao Qi’s Mengzi is not the only example of a text modeled on the Lunyu. Zheng Xuan’s 鄭玄 Hou Hanshu
biography (35.1212) also records that “his retainers together wrote up Zheng Xuan’s responses to his disciples’
questions about the Five Classics, and following the Lunyu composed the Mottos of Zheng in eight sections” (依論
語作鄭志八篇). 門人相與撰玄荅諸弟子問五經，依論語作鄭志八篇。).
319.Hou Hanshu 48.1599.
320.Baihu tongyi, “San jiao 三教” (SBCK 7/10a).
321.Xiaojing 7, “San cai 三才” (SBCK 1/5a).
152
Here as elsewhere, the Baihu tongyi compiler was quoting Kongzi without explicitly acknowledging that fact. By this time, the widespread view that Kongzi was ultimately responsible not
just for the Lunyu but also for the content of the Liji, Xiaojing, and other canonical texts meant
that the quotation of any one of these amounted to a quotation of Kongzi. Such precise, sourcebased attribution was only possible in a milieu of well-defined and widely circulating written
texts.
• Master (zi 子) texts
The Lunyu figures just as prominently in Kongzi quotations from Eastern Han master (zi 子)
texts. One of these was Wang Fu’s 王符 (c. 90–145 CE) Qianfu lun 潛夫論 (Discussions of a
Recluse), which explicitly quotes Kongzi in eighteen instances, twelve of which are direct quotations of the Lunyu. But the Lunyu’s influence is felt just as keenly in the implicit Lunyu quotations which pervade the text. The first chapter alone, entitled “Zan xue 贊學” (“In Praise of
Learning”) in apparent emulation of Lunyu book 1 (“Xue er 學而” [“Learning and”]), adds seven
implicit Lunyu quotations to its two explicit quotations. By Wang Fu’s time, the language and
ideas of the Lunyu permeated Eastern Han intellectual discourse.
One of the most valuable sources of Kongzi quotation practice is Wang Chong’s 王充 (27–
100) Lunheng 論衡 (Balanced Discourses). In addition to eight explicit Lunyu attributions,
Wang Chong included 115 Kongzi quotations in the Lunheng, 72 of which derived from the Lunyu. Ever the contrarian and critic, Wang Chong reveled in pointing out popular misconceptions,
in the course of which he revealed his attitudes about the trustworthiness of various texts and authorities. When it came to critiquing Kongzi sayings, no text was more authoritative for Wang
153
Chong than the Lunyu. Not only is the Lunyu the only source of Kongzi sayings to be explicitly
cited in the Lunheng, Wang Chong also used it to question the authenticity of non-Lunyu Kongzi
traditions like this one:
Some transmitted writings say that Yan Yuan and Kongzi together climbed Mt. Tai in Lu.
When Kongzi surveyed the southwestern view, there was a white horse tethered outside the
Wu palace gates [hundreds of miles distant]. He pointed this out to Yan Yuan, saying, “Do
you see the Wu palace gates?” Yan Yuan said, “I do.” Kongzi said, “What is outside the
gates?” “Something that looks like roped silk.” Kongzi rubbed his eyes and corrected himself,
whereupon they both came down the mountain. Afterwards, Yan Yuan’s hair turned white, his
teeth fell out, and he eventually died of illness. This was likely due to the fact that his vital
spirit could not compare to Kongzi’s. By exerting his strength to the utmost he exhausted his
vitality and thus died young.
Everybody nowadays who hears this believes it. But if you really assess it, it is surely an
empty saying. If you consult the text of the Lunyu, you will not find these words; if you
examine the traditions of the Six Classics, they also do not have this story.
傳書或言：顏淵與孔子俱上魯太山，孔子東南望，吳閶門外有繫白馬，引顏淵指以示
之，曰：若見吳昌門乎？顏淵曰：見之。孔子曰：門外何有？曰：有如繫練之狀。孔子
撫其目而正之，因與俱下。下而顏淵髮白齒落，遂以病死。蓋以精神不能若孔子，彊力
自極，精華竭盡，故早夭死。
世俗聞之，皆以為然。如實論之，殆虛言也。案論語之文，不見此言；考六經之
傳，亦無此語。322
This was not Wang Chong’s only objection to the Mt. Tai anecdote. In typical fashion, Wang
Chong went on to point out the story’s inherent implausibility. But the first test he subjected it to
was whether it was corroborated by the Lunyu.
Chapter 28 of the Lunheng, “Wen Kong 問孔” (“Interrogating Kongzi”), provides the clearest evidence of Wang Chong’s respect for the Lunyu. Aside from the Lunyu commentaries preserved in the Lunyu jijie, “Wen Kong” engages the Lunyu more directly than any other extant
text from the Eastern Han period. Perhaps because criticizing Kongzi and the Lunyu was controversial in his time, in the introduction to the chapter Wang Chong preemptively defended his intent to “criticize Kongzi” (nan Kongzi 難孔子) and “interrogate Kongzi’s words” (wen Kongzi
zhi yan 問孔子之言) by claiming that he simply followed “the way of interrogating and raising
322.Lunheng 16 (SBCK 4/2a–b; KZJY 16.3/pp. 360–61). For a similar example, see Lunheng 26 (SBCK 8/4b).
154
objections” (wen nan zhi dao 問難之道) in order to reveal the “truth” (shi fei 是非) of Kongzi’s
teachings. Without such questioning, he argued, the wisdom of Kongzi’s teachings would go unappreciated. He then proceeded to critique fourteen Kongzi sayings,323 all but one of which are
word-for-word quotations of the received Lunyu.324
One of Wang Chong’s targets was Lunyu 5/10:
Zai Wo fell asleep during the day. The Master said, “Rotten wood cannot be carved; dung soil
cannot be smoothed. What reprimand is there for Zai Yu?”
宰予晝寢。子曰：朽木不可雕也，糞土之牆不可杇也；於予與何誅？
As Wang Chong noted, the problem with the saying is its apparent contradiction with Lunyu
11/3, which includes “Zai Wo 宰我” (probably a variant name for Zai Yu 宰予) among those
Kongzi followers who were especially skilled at speaking. He then asked,
If Zai Wo’s character was as bad as rotten wood and dung soil, it wasn’t proper for him to
have entered Kongzi’s gates [as a student] or to be numbered among the four classes [of
exemplary disciples in Lunyu 13/3].
使宰我性不善，如朽木、糞土，不宜得入孔子之門，序在四科之列。325
Setting aside the question of whether Wang Chong’s critique is a valid one, Wang Chong clearly
expected a certain degree of consistency from Lunyu Kongzi sayings. The reference to Lunyu
11/3 in a discussion of Lunyu 5/10 shows that Wang Chong was thinking of the Lunyu as a single, ostensibly coherent text and not just as a source of discrete sayings. By implication, nonLunyu sayings did not accurately represent Kongzi’s teachings and so did not lend themselves to
this kind of analysis.
323.See chapter one, p. 88, for Wang Chong’s treatment of Lunyu 14/25.
324.The lone exception is a Kongzi saying from the “Tan gong 檀弓” (“Sandalwood bows”) chapter of the Liji. For
the Lunheng reference, see SBCK 9/13a. For the Liji original, see SBCK 2/9b; KZJY p. 520.
325.SBCK 9/5b.
155
• Bibliography
The earliest extant bibliography in the Chinese tradition, the Hanshu “Yiwen zhi 藝文志”
(“Monograph on Arts and Letters”), is an early Eastern Han version of Liu Xin’s 劉歆 (46 BCE–
23 CE) digest of an imperial catalogue initiated in 26 BCE by his father, Liu Xiang 劉向 (79–8
BCE), at the behest of Emperor Cheng 成帝.326 The complicated history of the “Yiwen zhi”
makes determining whether a given passage was penned by Liu Xiang in the Western Han, his
son in the Western Han or Xin, or Ban Gu 班固 (32–92 CE) in the Eastern Han next to impossible. But we will be on relatively safe ground if we read it as a summa of early Eastern Han
knowledge about received textual traditions.
The first division of the “Yiwen zhi,” titled “Liu yi 六藝” (“The Six Arts”), relates the history of the most prestigious and authoritative texts in the period and is arranged in the following
order: the Yi 易 (Changes), Shu 書 (Documents), Shi 詩 (Odes), Li 禮 (Ritual), Yue 樂 (Music),
Chunqiu 春秋 (Spring and Autumn Annals), Lunyu, Xiaojing 孝經 (Classic of Filial Piety), and
various elementary learning (xiao xue 小學) texts. There is certainly nothing remarkable about
the placement of the Five Classics plus the Music classic atop this list. What is remarkable is the
Lunyu’s inclusion immediately after these texts, presumably because the Lunyu was perceived as
the most authoritative text after the Six Classics. References to the “Seven Classics” (qi jing 七
經), i.e., the Six Classics plus the Lunyu, appear first in texts dated to the Eastern Han.327 Perhaps
this honorific had something to do with Ban Gu’s (or Liu Xiang’s or Liu Xin’s) catalogue.
326.Liu Xiang’s text was called the Bie lu 別錄 (Catalogue) and Liu Xin’s Qi lüe 七略 (Seven Summaries). Yan
Kejun collected the extant fragments of the Bie lu and Qi lüe in the Quan Han wen (38/1a–8b, pp. 336–339, and
41/4b–7a, pp. 351–53).
327.See, e.g., Hou Hanshu 35.1196 and Quan Hou Han wen 75, p. 881a.
156
Judging by the rhetoric of the “Yiwen zhi” and the kinds of texts invoked in its historical survey of early textual culture, the Lunyu’s rhetorical value was rivaled only by that of the Zhouyi,
no doubt thanks to Kongzi’s perceived role as the author, compiler, and editor of the Five Classics.328 As the opening lines of the “Yiwen zhi” state:
Formerly, after Zhongni passed away his subtle words were cut off, and after the seventy
disciples died the greater meaning [of his teachings] was distorted. Thus the Chunqiu was split
into five [traditions], the Shi into four [traditions], and the Yi into traditions of several experts.
Amid the strategic scheming of the Warring States the true and the counterfeit came into
conflict and the words of the various masters became chaotic and confused.
昔仲尼沒而微言絕，七十子喪而大義乖。故春秋分為五，詩分為四，易有數家 之傳。
戰國從衡，真偽分爭，諸子之言紛然殽亂 。329
Note the strong claim implied by the use of the connective gu 故 (thus). Confusion about canonical texts arises from a confusion about Kongzi’s original teachings, and ignorance of those teachings creates opportunities for inferior “experts” (jia 家) and “masters” (zi 子). According to this
interpretation of early intellectual history, the key to resolving confusions about the canon is
recovering Kongzi’s “subtle words” and “greater meaning.” If the dozen Lunyu quotations in the
text are any indication, the Lunyu filled this need more directly than the coded language of the
Chunqiu or Shi and more authoritatively than the Kongzi sayings of the Liji.
Like the Lunyu jijie, the “Yiwen zhi” records a number of Lunyu companion texts that together attest to scholarly interest in the Lunyu (see Appendix 2:D). The particulars of the “Yiwen
zhi” account need not concern us at the moment. What is relevant for the present discussion is
that none of the Lunyu transmitters and commentators named in the entry pre-date the reign of
Emperor Wu 武帝 (r. 141–87 BCE) (see Appendix 2:E).330 One wonders if the “Yiwen zhi” did
328.By my count, the “Yiwen zhi” quotes the Lunyu and Zhouyi a dozen times apiece, the Xiaojing and Shu x3, and
li 禮 texts x2.
329.Hanshu 30.1701.
330.For this observation, see Zhao Zhenxin 1961, p. 24, and Makeham 1996, p. 19. Zhao Zhenxin concludes from
157
not list earlier Lunyu experts because there was no Lunyu to be an expert in.
Regardless, the Hanshu “Yiwen zhi” is not the only source of Eastern Han Lunyu bibliography. In chapter 81 of the Lunheng, “Zheng shuo 正說” (“Rectifying explanations”), Wang
Chong criticizes popular misconceptions about the Five Classics as well as the Lunyu and
Mengzi (see Appendix 2:F). In its broad outlines, Wang Chong’s polemical account of the Lunyu’s history does not depart significantly from the “Yiwen zhi.” One key difference has to do
with textual loss, a theme which grounds Wang Chong’s criticisms of contemporary Lunyu commentators. He complained that scholars of his era did not understand why their received Lunyu
looks the way it does because earlier recensions—the Qi 齊, Lu 魯, and Hejian 河間 versions–
were no longer extant. Scholars were equally ignorant of the Lunyu’s physical format in the preHan era due to its having been lost prior to its rediscovery during the reign of Emperor Wu (141–
89 BCE). How Wang Chong arrived at these views is a mystery—perhaps it reflects a bibliographic tradition which did not find its way into the Liu Xiang–Liu Xin–Ban Gu catalogues.
Nonetheless, it is striking to find an Eastern Han author in a decidedly post-Lunyu milieu pointing out significant gaps in the history of the Lunyu text, gaps which happen to coincide with the
big picture of Kongzi quotation practice outlined above.331
The Xin period (9–23 CE)
this observation that the Lunyu was first compiled during Emperor Wen or Jing’s reign.
331.Zhao Qi’s preface to his Mengzi commentary also happens to include the statement that a boshi 博士 (erudite)
position was established for the Lunyu and the Mengzi by Emperor Wen 文帝 (r. 180–157). However, the fact that
this claim is not corroborated by any earlier source suggests to me that it reflects an Eastern Han as opposed to an
early Western Han conception of the Lunyu’s authority. As we have seen, Zhao Qi sought to present his Mengzi as a
Lunyu copycat text, so it is easy to imagine Zhao Qi inventing such a story in order to cement the relationship
between the texts. On this point see also Kaneto 1980, pp. 21–22.
158
Lunyu Kongzi sayings featured just as prominently in the rhetoric leading up to Wang Mang’s
(46 BCE–23 CE) overthrow of the Liu clan and his establishment of the short-lived Xin 新
dynasty (9–23 CE) as they did in the Cao clan’s establishment of the Wei dynasty. Much like
Cao Pi’s officials in the month before his accession, Wang Mang’s officials repeatedly extolled
his virtues in the years leading up to 9 CE. One of these men was Chen Chong 陳崇, a high official who celebrated Wang Mang in a 3 CE memorial punctuated with quotations from the Shi,
Shu, Yi, and especially the Lunyu:332
...[Lunyu 1/15] Kongzi said, “It is better to be poor yet joyful, rich yet loving ritual.” This
refers to the Duke [i.e., Wang Mang].
...孔子曰：未若貧而樂，富而好禮，公之謂矣。
...[Lunyu 17/6] Kongzi said, “If he acts with alacrity he will produce results.” This refers to the
Duke.
...孔子曰：敏則有功，公之謂矣。
...[Lunyu 4/13] Kongzi said, “Being able to run a state with ritual and deference—what
[problem] can there be with that?” This refers to the Duke.
...孔子曰：能以禮讓為國乎何有，公之謂也。
...[Lunyu 1/14] Kongzi said, “He eats without seeking to be full; he dwells without seeking
rest.” This refers to the Duke.
孔子曰：食無求飽，居無求安，公之謂矣。
Through his Lunyu Kongzi quotations, Chen Chong invited his audience to think of Wang Mang
as the incarnation of the Lunyu Kongzi’s ideal ruler.
Wang Mang himself appealed to the Lunyu Kongzi on at least two occasions. On January
6th, 9 CE, Wang Mang issued a memorial to his aunt, the Grand Empress Dowager, requesting
that he be acknowledged as “Emperor” (huangdi 皇帝) in his day-to-day management of the
332.Hanshu 99a.4053–4063. The Hanshu records that this memorial was drafted by Zhang Song 張竦 (d. 23 CE).
159
realm. A quotation of Lunyu 16/8, which was also invoked by Cao Pi’s highest officials in 220,333
anchored Wang Mang’s arrogation of imperial authority to a widely recognized Kongzi saying:
Kongzi said, “Fear the Mandate of Heaven, fear the great, and fear the sayings of the sages.”
How can your servant [Wang] Mang presume not to obey? Your servant begs that when
respectfully serving the spirits and the ancestral temples and when memorializing to the Grand
Empress Dowager and the Filial Emperor Ping he be allowed to call himself “the Acting
Emperor,” [but] when announcing proclamations or ordinances to the empire or when
receiving memorials from the realm he should not be called “Regent” [and should simply be
called “Emperor”].
孔子曰：畏天命，畏大人，畏聖人之言。臣莽敢不承用！臣請共事神祇宗廟，奏言太皇
太后、孝平皇后，皆稱假皇帝。其號令天下，天下奏言事，毋言攝。334
After becoming emperor, Wang Mang quoted Kongzi again in 20 CE after a “great wind”
(dafeng 大風) had destroyed a palace building, an inauspicious event which led Wang Mang to
admit to various failures of his rule. One of these was a failure to “correct names” (zheng ming
正名) in keeping with the advice at Lunyu 13/3:
“[My son Wang] Lin is called the Crown Prince even though he has an older brother—this is a
case of names not being correct. His Excellency the Expansive Ni [=Kongzi] said, “If names
are not correct then speech does not accord [with reality],” and this amounts to “punishments
not being appropriate and the people having nowhere to place their hands and feet.”
臨有兄而稱太子，名不正。宣尼公曰：名不正，則言不順，至於刑罰不中，民無錯手
足。335
“His Excellency the Expansive Ni” (Xuan Ni gong 宣尼公) is an abbreviated form of “His Excellency the Expansive Ni of Baocheng” (Baocheng Xuan Ni gong 褒成宣尼公), a posthumous
title (shi 諡) bestowed on Kongzi in 1 CE by the puppet Emperor Ping 平帝 (r. 1 BCE–6 CE).336
Presumably, Wang Mang’s use of this honorific was meant to communicate his profound respect
for Kongzi and everything he represented. However, later rulers did not adopt the convention,
333.Sanguo zhi 2.73.
334.Translation adapted from Dubs 1955, 2.251–2. Wang Mang concluded his memorial with a promise to
relinquish authority to the young emperor Liu Ying 劉嬰 (r. 6–8 CE).
335.Hanshu 99c.4160.
336.Hanshu 12.351.
160
perhaps because of its association with Wang Mang.
Wang Mang’s use of Lunyu 13/3 also communicates his respect for Kongzi by demonstrating
his commitment to the precise text of the Lunyu. The relevant section of Lunyu 13/3 reads as
follows:
The Master said, “What a rube you are, Zilu! A noble man when dealing with matters he does
not understand should keep silent. When names are not correct speech does not accord; when
speech does not accord affairs are not completed; when affairs are not completed ritual and
music do not flourish; when ritual and music do not flourish punishments are not appropriate;
when punishments are not appropriate the people have nowhere to place their hands and feet.
A noble man is never lax about his speech.”
子曰：野哉由也！君子於其所不知，蓋闕如也。名不正，則言不順；言不順，則事不
成；事不成，則禮樂不興；禮樂不興，則刑罰不中；刑罰不中，則民無所錯手足。故君
子名之必可言也，言之必可行也。君子於其言，無所苟而已矣。
Rather than quote the entirety of one of the longest Kongzi sayings in the Lunyu, Wang Mang
quoted only the initial and final sections (highlighted in dark gray) of the causal chain beginning
with “when names are not correct” (ming bu zheng 名不正). He then abbreviated the intervening
links (highlighted in light gray) with the phrase “and this amounts to” (zhi yu 至於). Although
one might be tempted to take the variation as evidence that Wang Mang was quoting a Lunyu
variant and not the Lunyu itself, in fact the form of the quotation seems to indicate a high degree
of fidelity to the Lunyu text. Apparently, Wang Mang (or the official who drafted the document)
was loathe to pick and choose phrases from Lunyu 13/3 without acknowledging the omissions.
Wang Mang’s last words as reported in his Hanshu biography, like so many famous last
words, should be interpreted more from a literary than a historical perspective. Regardless of
their provenance, it is significant that they were inspired by Lunyu 7/23: as the Han armies
closed in on his position, Wang Mang supposedly exclaimed, “Heaven birthed the virtue in me,
so what can the Han army do to me?” ([王莽]曰：天生德於予，漢兵其如予何！).337 In the ap-
337.Hanshu 99c.4190. Lunyu 7/23 reads: “The Master said, ‘Heaven birthed the virtue in me, so what can Huan Tui
161
praisal (zan 贊) which concludes his Wang Mang biography, Ban Gu 班固 (32–92 CE) went on
to write,
Formerly, the Qin burned the Shi and Shu in order to sow selfishness and dissension, and
Wang Mang chanted the Six Arts in order to embellish treacherous speech. They both
“returned to the same place but along different paths” and thereby met with destruction.338
昔秦燔詩書以立私議，莽誦六藝以文姦言，同歸誅塗，俱用滅亡。339
In Ban Gu’s mind, Wang Mang’s use of Lunyu 7/23 seems to have represented yet another instance of a usurper “chanting the Six Arts in order to embellish treacherous speech.” The choice
of a Lunyu allusion as opposed to an allusion to one of the Five Classics is yet another indication
of the Lunyu’s symbolic authority in the period, one that recalls Zhao Qi’s description of the text
as “the linch-pin of the Five Classics and the mouthpiece of the Six Arts” (五經之錧鎋，六藝之
喉衿也。).340 Although not initially considered a classic (jing 經) in its own right, as the pre-eminent source of Kongzi sayings it provided direct access to the mind which had created the
classics.
The late Western Han period (100 BCE–9 CE)
• Yang Xiong 揚雄 (53 BCE–18 CE)
Even though its single explicit Kongzi attribution makes it a poor source of Kongzi quotation
do to me?” (子曰：天生德於予，桓魋其如予何？). According to the “Kongzi shijia” (Shiji 47.1921), Huan Tui
tried to kill Kongzi on his travels to the state of Song 宋.
338.The line “returned to the same place but along different paths” is a quotation from the Zhouyi “Xici zhuan xia
繫辭傳下” (SBCK 8/3b).
339.Hanshu 99c.4194.
340.For this passage, see p. 151 above.
162
practice, Yang Xiong’s 揚雄 (53 BCE–18 CE) Fayan 法言 (Model Sayings) testifies to the Lunyu’s importance as a model for authors in the period. According to Yang Xiong’s Hanshu
biography,
Yang Xiong saw that the various masters each with his own wisdom railed against the other,
and that the bulk of them slandered the sages. They engaged in the strange and the esoteric,
they split hairs with their disputations and beguiled with their phrases and thereby disrupted
contemporary affairs. Although the scale of their disputations was small, in the end they
destroyed the great Way and confused the masses by drowning them in hearsay and making
them unaware of their own errors. Even His Excellency the Grand Scribe [i.e., Sima Qian] in
his record of the Six Kingdoms down to Chu and Han stopped with the [capture of the] lin
[just as Kongzi did in the Chunqiu]. But his record was not in agreement with the sages, and
his morality strayed to a certain extent from the classics. And so when people in his day asked
Yang Xiong questions he often used [the sages’] models in reply. These he compiled into a
thirteen-chapter text which he called the Model Sayings in imitation of the Lunyu.
雄見諸子各以其知舛馳，大氐詆訾聖人，即為怪迂，析辯詭辭，以撓世事，雖小辯，終
破大道而或眾，使溺於所聞而不自知其非也。及太史公記六國，歷楚漢，（記）〔訖〕
麟止，不與聖人同，是非頗謬於經。故人時有問雄者，常用法應之，譔以為十三卷，象
論語，號曰法言。341
What did it mean for the Fayan to have been written “in imitation of” (xiang 象) the Lunyu?
Consider the opening lines of Fayan book one, which is itself entitled “Xue xing 學行”
(“Learning in practice”) in apparent emulation of Lunyu book one, “Xue er 學而” (“Learning
and”):
To learn and practice [what one has learned] is best. To speak of [what one has learned] is
second best, and to teach others [what one has learned] is next best. The masses can do none
of these things.
學，行之，上也；言之，次也；教人，又其次也。咸無焉，為眾人。342
While Yang Xiong quotes neither Kongzi nor the Lunyu in this passage, an audience trained in
the Lunyu would not have missed the homage to Lunyu 16/9:
Kongzi said, “To know something from birth is best. To know it by learning is second best. To
have learned it after experiencing some difficulty is next best. To have experienced difficulty
but not to have learned anything—the people who do this are the worst.
孔子曰：生而知之者上也，學而知之者次也；困而學之，又其次也；困而不學，民斯為
341.Hanshu 87.3580.
342.SBCK 1/1a.
163
下矣。
What Yang Xiong borrowed from Lunyu 16/9 was not its content but its structure. In Yang
Xiong’s hands, Lunyu 16/9 became a template for producing new sayings based on the formula
“X is best; Y is next best; and Z is the worst.” Quoting Lunyu 16/9 word-for-word would have
been derivative; treating it as a model (fa 法) signaled his respect for Kongzi while also giving
Yang Xiong the freedom to create a new text of his own. The other respect in which Yang Xiong
imitated the Lunyu was in the compactness of the Fayan’s language. The Fayan version removes
extraneous conjunctions (e.g., er 而), nominalizing particles (zhe 者), and object pronouns (zhi
之) in an effort to maximize the line’s pithiness. Presumably, such compression lent the text an
archaic, or perhaps a transparently archaicized, air.343
• Liu Xiang 劉向 (79–8 BCE)
As the official charged in 26 BCE with collating and cataloging the imperial collection, Liu Xiang 劉向 (79–8 BCE) was in a position to shape the transmission of early textual traditions more
profoundly than any other figure from the period. Earlier I noted that Liu Xiang compiled the
first imperial catalogue, the Bie lü 別錄 (Catalogue), which survives in abridged form as the
“Yiwen zhi” (as well as in a few fragments). A prolific compiler, Liu Xiang was also responsible
for three significant compendia: the Shuiyuan 說苑 (Garden of Persuasions), Xinxu 新序 (New
Arrangement), and Gu lienü zhuan 古列女傳 (Biographies of Ancient Women).
The largest of these was the Shuiyuan, a text of 107,000 characters with a total of 194 total
343.On the language of the Fayan, see Schilling 2006.
164
Kongzi attributions. Despite the likelihood that Liu Xiang was ultimately responsible for the
Lunyu’s privileged place in the imperial catalogue, only twenty-four (13%) of the Shuiyuan’s
Kongzi attributions have parallels in the Lunyu, a much smaller percentage than for any text discussed in this chapter thus far. Unlike, say, Wang Chong, Liu Xiang did not rely exclusively on
the Lunyu for his Kongzi quotations, nor did he explicitly privilege the twenty-four Kongzi sayings with Lunyu parallels sayings over the 160 Kongzi sayings without.
However, Liu Xiang’s handling of Lunyu Kongzi sayings does suggest that he valued them
more than Kongzi sayings from other sources.344 For instance, only Lunyu Kongzi sayings (with
one exception)345 function as exordia, i.e., as jumping-off points for expositions of various sorts.
Such quotations are typically followed by a statement prefaced by the particle fu 夫, which
marks the general principle or lesson to be drawn from the quotation.346 In addition, only Lunyu
Kongzi sayings are explained or commented on, e.g., when a quotation of Lunyu 12/19 (Kongzi
said, “The virtue of the noble man is wind and that of the petty man is grass. Grass in wind always bends” [君子之德，風也；小人之德，草也；草上之風必偃。]) is followed by the
comment, “this is to say that [the noble man] merely makes his transformative influence shine
forth” (言明其化而已矣).347 Apparently, Liu Xiang did not believe that non-Lunyu Kongzi sayings deserved such close attention.
In one passage from Shuiyuan book 19, “Xiu wen 修文” (“Cultivating Patterning”), Liu Xi-
344.This observation is also an important reminder that the quantitative approach to Kongzi quotation practice only
takes us so far. In my experience, we learn just as much, if not more, about the relative authority of Lunyu Kongzi
sayings from the ways that early authors handled this material.
345.This is the saying “Kongzi said, ‘I follow indirect remonstrance’” (孔子曰：吾其從諷諫乎。) from book nine,
“Zheng jian 正諫” (“Correct remonstrance”). See SBCK 9/1b.
346.For this usage of fu 夫, see Graham 1972, pp. 107–108..
347.Shuiyuan 7 (SBCK 7/1b).
165
ang even provides a full-fledged commentary on Lunyu 6/2 (see Appendix 2:G for a translation).
The received text of Lunyu 6/2 is introduced with a question from one of Kongzi’s disciples:
“Zhonggong asked about Zisang Bozi. The Master said, ‘He is acceptable for his simplicity’” (仲
弓問子桑伯子。子曰：可也簡。). Kongzi’s comment in the Lunyu is then followed by a saying attributed to Zhonggong on the subject of jian 簡 (simplicity), and the entire entry ends with
Kongzi expressing his approval of Zhonggong (“Yong’s [=Zhonggong’s] words are right” 雍之
言然). But the Lunyu provides no background whatsoever about Zisang Bozi, the situation that
prompted Kongzi’s evaluation, or the relevance of the Zhonggong saying to Zisang Bozi aside
from the shared use of jian 簡. Without this background, the logic of the Lunyu saying remains
entirely opaque.348
Liu Xiang stepped into this information vacuum with his discussion of Lunyu 6/2. After glossing jian 簡 (simplicity) as yi ye 易野 (unsophistication), he relates a Kongzi anecdote that provides the backstory for Lunyu 6/2:
Kongzi went to see Zisang Bozi, who was sitting in his residence without a cap. A disciple
said, “Master, why would you see such a man?” [Kongzi] said, “His substance is fine though
he is unrefined. I wish to persuade him to refine himself.” After Kongzi left Zisang Bozi’s
retainers were unhappy and said, “Why would you see Kongzi?” [Zisang Bozi] said, “His
substance is fine though he is excessively refined. I wish to persuade him to unrefine himself.”
孔子見子桑伯子，子桑伯子不衣冠而處，弟子曰：夫子何為見此人乎？曰：其質美而無
文，吾欲說而文之。孔子去，子桑伯子門人不說，曰：何為見孔子乎？曰：其質美而文
繁，吾欲說而去其文。349
348.See Weingarten 2010, pp. 151–154, for a parallel discussion of this passage. He concludes that “the Shuoyuan
passage was based on a text in which the passages corresponding to Lunyu paragraphs 6/1 and 6/2 already followed
one another.” His main evidence for thinking that Liu Xiang was quoting a Lunyu source text and not the Lunyu
itself is the presence of a Lunyu 6/18 variant (“One who cultivates his patterning and substance is called a ‘noble
man,’ and one who has substance but no patterning is called a ‘simple rustic.’”) prefaced by gu yue 故曰 (thus it is
said).
349.It is curious that this episode gives Zisang Bozi the final word with a mirror-image evaluation that reveals the
one-sidedness of Kongzi’s perspective. This leads me to suspect that the original version of this episode was not
designed to enhance Kongzi’s image, but that Liu Xiang framed it as a critique of Zisang Bozi in order to explain
Lunyu 6/2.
166
In a comment immediately following the anecdote, the Shuiyuan labels someone like Kongzi
who “cultivates his patterning and substance” (wen zhi xiu zhe 文質修者) a noble man (junzi 君
子), whereas someone like Zisang Bozi who rejects “cultivation” (xiu 修) is a “simple rustic” (yi
ye 易野). Earlier we saw that Liu Xiang went out of his way to declare Kongzi the sole winner of
the “man of Chu lost his bow” episode by cutting out Laozi altogether (p. 57). Given this interest
in presenting Kongzi as an unrivaled authority, it is curious that the present anecdote gives
Zisang Bozi the last word without any comment or criticism from Kongzi. The juxtaposition of
Kongzi’s evaluation with that of Zisang Bozi seems to highlight the one-sidedness of the masters’ perspectives, as if both represented two sides of the same coin. This is a decidedly less gong
公 (impartial) Kongzi than observed in other texts
In any event, in the conclusion of the section Liu Xiang goes on to point out that
Zhonggong’s perspicacity in Lunyu 6/2 explains Kongzi’s praise of Zhonggong at Lunyu 6/1
(“The Master said, ‘Yong [=Zhonggong] can be made to face south” [子曰：雍也可使南面。]).
Like Wang Chong, Liu Xiang used one Lunyu Kongzi saying to comment on another, presumably because he was used to thinking of the Lunyu as a single, internally consistent source. At the
same time, his willingness to use a non-Lunyu Kongzi anecdote to explain the genesis of a Lunyu
Kongzi saying reveals an openness towards non-Lunyu Kongzi material that we did not observe
in, say, Wang Chong’s writings. Liu Xiang’s more accommodating approach is one of the first
signs that the Lunyu’s pre-eminence in the Eastern Han did not necessarily extend to earlier
periods.
• Edicts
167
Kongzi attributions in late Western Han edicts look much the same as they do in the Xin and
Eastern Han insofar as late Western Han emperors invoked the Lunyu Kongzi roughly as often as
their Eastern Han successors. However, the picture begins to look very different as we move beyond the reign of Emperor Yuan 元帝 (r. 49–33 BCE) (see Appendix 2:H), the first emperor we
know of who used the words Kongzi yue 孔子曰 (Kongzi said) in an extant imperial communication. That document, an admonishing letter to Liu Yu 劉宇, the wayward King of Dongping 東
平王 (d. 20 BCE), concludes with the most menacing Kongzi quotation in the early corpus:
Kongzi said, “To err without reforming is called a [true] error” [Lunyu 15/30]. You, King,
must deeply consider and thoroughly reflect on this. Do not go against Our will.”
孔子曰：過而不改，是謂過矣。王其深維孰思之，無違朕意。350
The distribution of Kongzi quotations in Western Han edicts happens to coincide with what
Michael Loewe has described as an “age of transition” in which the “modernist” policies of Emperor Wu’s reign gave way to “reformist” policies during the reigns of Emperors Zhao 昭帝 (87–
74 BCE) and Xuan 宣帝 (74–49 BCE).351 In their effort to scale back the excesses of Emperor
Wu’s reign, late Western Han reformists frequently invoked Zhou 周 dynasty precedents, a
rhetorical strategy eventually co-opted by Wang Mang in spectacular fashion in the run-up to the
Xin dynasty. Writing of roughly the same period, Martin Kern has also noted the advent of a “ritual classicism” anchored to “the texts of the gradually emerging state-sponsored canon.”352 For
reformist politicians like Kuang Heng 匡衡 (d. 30/29 BCE),353 this classicism was an antidote to
the elaborate state rituals instituted during Emperor Wu’s reign, rituals whose ideology derived
350.Hanshu 80.3321.
351.Loewe 1986, pp. 185ff.
352.Kern 2001, p. 66. The lens through which Kern discusses these issues is a shift in the meaning of wenzhang 文
章 from ritual practice to ritual canon.
353.For a discussion of Kuang Heng’s role in promoting this classicism, see Kern 2001, pp. 61–73.
168
from non-canonical sources. (Fukui Shigemasa’s argument that the earliest reliable mention of
the term “The Five Classics” [wu jing 五經] dates to 51 BCE, not the early part of Emperor
Wu’s reign as is usually supposed, is relevant in this context.)354 Considered from this perspective, the uptick in imperial Kongzi quotations beginning in Emperor Yuan’s reign is an index of
larger trends in Western Han political culture. Kongzi was a natural ally for the late Western Han
statesmen and scholars who sought to build a political program atop the Five Classics.
The first explicit mention of Kongzi in an edict dates to 123 BCE, eighteen years into Emperor Wu’s 武帝 reign:
We have heard that the Five Emperors did not continue their predecessors’ rituals and the
three dynasties did not share the same models. Although they followed different paths the
virtue they established was the same. Kongzi replied to Duke Ding that he should attract the
distant, to Duke Ai that he should assess his subjects, and to Duke Jing that he should
moderate his expenses. It was not that their desires were different, it was that their problems
demanded different solutions.
朕聞五帝不相復禮，三代不同法，所繇殊路而建德一也。蓋孔子對定公以徠遠，哀公以
論臣，景公以節用，非期不同，所急異務也。355
It is possible that the edict’s Kongzi quotation derived from Lunyu 13/16, an exchange between
Kongzi and the Duke of She in which Kongzi advises the Duke to “attract the distant” (yuanzhe
lai 遠者來). However, nowhere in the Lunyu does Kongzi urge a lord to “assess subjects” (lun
chen 論臣) or “moderate expenses” (jie yong 節用) as he is said to here.356 In fact, a much better
match for Emperor Wu’s Kongzi attribution is a passage from book 38 of the received Han Feizi,
“Nan san 難三” (“Objections, part 3”) (see Appendix 2:I). The Han Feizi version is not a perfect
parallel because it differs with respect to the first interlocutor (Zigao versus Duke Ding) and its
advice to Duke Ai (xuan xian 選賢 [selecting the worthy] versus lun chen 論臣 [assessing sub-
354.Fukui 1994, 1995, & 1997.
355.Hanshu 6.173. Translation adapted from Dubs vol. 2, pp. 55–56.
356.Lunyu 1/5 uses the phrase jie yong 節用 but not in reference to a lord.
169
jects]). Whether or not the Han Feizi was the actual source of Emperor Wu’s Kongzi attribution,
the comparison with the Han Feizi makes it much harder to read the edict as a quotation of the
received Lunyu.
Three additional edicts, one by Emperor Xuan in 66 and two more by Emperor Wu in 112
and 128, include unattributed parallels which might be interpreted as Lunyu quotations:
<Emperor Xuan in 66 BCE>A tradition says, “Filial piety and brotherly respect are surely the
roots of humaneness.”
傳曰：孝弟也者，其為仁之本與。357
<Emperor Wu in 112 BCE>We have heard to repay virtue with virtue and to repay wrong
with rectitude.
朕聞報德以德，報怨以直。358
<Emperor Wu in 128 BCE>In a village of ten houses there is always [someone who is] loyal
and trustworthy. Among three men traveling together there will be one who can be my
teacher.”
夫十室之邑，必有忠信；三人並行，厥有我師。359
The zhuan yue 傳曰 (a tradition says) attribution in Emperor Xuan’s edict is a close parallel to
Lunyu 1/2, Emperor Wu’s first edict opens with a variant of Lunyu 14/34, and the fu 夫 particle
in Emperor Wu’s second edict prefaces a two-part statement which combines parts of Lunyu 5/28
and Lunyu 7/22. Without more specific quotation markers it is impossible to tell whether Emperors Xuan and Wu (or, more accurately, the officials who drafted their edicts) intended these passages to be recognized as Lunyu quotations. But even if we accept these passages as legitimate
Lunyu quotations, the first sign of the Lunyu’s influence on Han imperial rhetoric would date no
earlier than 128 BCE.
357.Hanshu 8.250
358.Hanshu 58.2627.
359.Hanshu 6.166.
170
The earliest explicit mention of the Lunyu by a Han emperor, and one of the earliest descriptions of anyone actually using a “Lunyu” text, is this edict issued in 82 BCE by Emperor Zhao
when he was only thirteen years old:
We with Our insignificant person have obtained [the opportunity] to protect the [imperial]
ancestral temples. Tremblingly and circumspectly we have risen early and gone to bed late [in
order to] cultivate [Ourself] in the practices of the ancient lords and kings. [Although We]
have been made acquainted with the Xiaojing, the Lunyu, and Shangshu through the teaching
of [Our Grand] Guardian and [Grand] Tutor, [We can]not say that [We] have any perfect
understanding [of them].
朕以眇身，獲保宗廟，戰戰栗栗，夙興夜寐，修古帝王之事，通保傅，傳孝經、論語、
尚書，未云有明。其令三輔太常擧賢良各二人，郡國文學高第各一人，賜中二千石以下
至吏民爵各有差。360
The Hanshu elsewhere records that Emperor Zhao’s father, Emperor Wu, had selected a prince
in 91 BCE to serve as King of Guangchuan 廣川, his recorded qualifications including that he
had “received instruction in the Yi, Lunyu, and Xiaojing, all of which he had mastered” (師受
易、論語、孝經皆通).361 For emperors in this period the Lunyu was much more than a source
of memorable quotations. By Emperor Zhao’s reign a text called “Lunyu” was being used as a
textbook for the education of the Han royal family, and its value as a textbook was apparently so
well recognized that it could be invoked as a symbol of studiousness and virtue alongside the
Shangshu or Zhouyi.
Such symbolism mattered especially to Emperor Zhao and his handlers no doubt due to the
succession crises which marked the last few years of Emperor Wu’s reign. Liu Fuling 劉弗陵,
the future Emperor Zhao, was designated crown prince only after Liu Ju 劉據 (128–91), Emperor Wu’s long-time heir apparent, died in fighting which had broken out between rival court fac-
360.Hanshu 7.223. Translation adapted from Dubs 1955, 2.159–60.
361.Hanshu 53.2428. The prince in question was Liu Qu 劉去. However, instruction in the Lunyu was no guarantee
of virtue: “Few figures in Han history have been reported to have been so evil minded as Liu Qu” (Loewe 2000, p.
348).
171
tions in 91 BCE.362 The disgruntled prince Liu Dan 劉旦, an older son of Emperor Wu, went so
far as to question Emperor Wu’s paternity of Liu Fuling and even planned a coup d’état in 86
BCE. He survived the plot’s discovery only to stage another failed coup in 80 BCE, after which
he committed suicide.363 In 82 BCE, a man claiming to be Liu Ju 劉據, the former heir apparent,
presented himself at the palace gates and was subsequently executed.364 In 78 BCE, a memorial
submitted by an official named Sui Hong 眭弘 urged the young Emperor Zhao to yield the
throne to a true xianren 賢人 (man of accomplishment). Sui Hong was promptly executed for the
suggestion.365 Given this turmoil, Emperor Zhao’s handlers may have felt some pressure to justify Liu Fuling’s selection as emperor as well as their own stewardship.
Significantly, the Lunyu’s use as a textbook for the royal family continued through the end of
Western Han and beyond. As Mark Csikszentmihalyi has observed, “the fact that both authors of
commentaries to the Analects in the first century BCE and four of the text’s six experts [mentioned in the “Yiwen zhi”] served as grand tutors to the Heir Apparent shows that the text was
closely associated with the education of minors.”366 One of these experts was Xiahou Sheng 夏侯
勝, who composed a Lunyu shuo 論語說 (Explanations of the Lunyu) text during the early part
of Emperor Xuan’s 宣帝 reign while serving as Grand Tutor (taifu 太傅) to the crown prince.367
362.For a summary of this episode, see the entry for Liu Ju in Loewe 2000, pp. 321–322.
363.For Liu Dan’s biography, see Loewe 2000, p. 287–89.
364.For a summary of this episode, see the entry on Cheng Fangsui 成方遂 in Loewe 2000, p. 45.
365.Hanshu 75.3153–54. For Sui Hong’s (aka Sui Meng 眭孟) biography, see Loewe 2000, p. 496. The immediate
stimuli for Sui Hong’s memorial were a series of strange and inauspicious phenomena
366.Csikszentmihalyi 2002, p. 146. Csikszentmihalyi also connects this development with the rise of Han book
culture (p. 148): “Although books had economic value to individuals, the collection of books by the state likely
reflected their increasing educational value as, among other things, textbooks for use in educating both the Heir
Apparent and prospective civil servants.”
367.Hanshu 75.3159.
172
Two additional Lunyu experts, Xiao Wangzhi 蕭望之 (ca. 107–47) and Wei Xuancheng 韋玄成
(d. 36), served as Grand Tutors from 59–49 and 49–43, respectively. The Hanshu biography of
Zhang Yu 張禹 records that he wrote a Lunyu zhangju 論語章句 (chapter-and-verse Lunyu commentary) for the benefit of Liu Ao 劉驁, the future Emperor Cheng 成帝, who had reportedly
asked him numerous questions about the text.368 To this list one might also add the Eastern Han
scholar and Lunyu commentator Bao Xian 包咸 (8 BCE–65 CE), who taught the Lunyu to the
Crown Princes Liu Qiang 劉彊 (25–58 CE) and Liu Zhuang (28–75), the future Emperor Ming
明帝. Bao Xian’s son Bao Fu 包福 likewise tutored the young Emperor He 和帝 (r. 88–105).369
The discovery of the Dingzhou Lunyu manuscript from the tomb of Liu Xiu 劉脩 (d. 55 BCE),
the king of Zhongshan 中山, further testifies to the Lunyu’s circulation within the royal clan in
the latter part of the Western Han.370
• Memorials
While imperial edicts by no means tell the whole story of Kongzi quotation practice among the
368.Hanshu 81.3352.
369.Hou Hanshu 79b.2570. The Hou Hanshu only mentions these two figures because their tutors were also wellknown scholars. Elsewhere in the Hou Hanshu, the Empress (and later Empress Dowager) Deng Sui 鄧綏 (81–121)
is described as having “mastered” (tong 通) the Lunyu by the age of 12. The Empress (and later Empress Dowager)
Liang Na 梁妠 (116–150) was reported to have mastered it by the age of nine (Hou Hanshu 10b.438).
370.The Hanshu (1b.76) also records that Emperor Gaozu 高祖 (r. 202–195 BCE), the Han founder, performed the
tailao 太牢 sacrifice to Kongzi in 195 BCE while passing through the state of Lu. However, as Nylan and Wilson
(2010, pp. 71–72) have argued, “[i]f he did so—this is not an event inserted into the historical record until centuries
after 195 BCE—it seems unlikely that Gaozu intended by this act to signal a special alignment between his ruling
house and Kongzi, since the standard histories for Han show a throne intent upon establishing its ritual supremacy in
the customary way, by supporting and co-opting as many local cults as it could reach out to. Quite inexplicable, if
we buy the standard line, are two curious facts: (a) no descendant of Kongzi was awarded high rank purely on the
basis of his descent from Kongzi until 1 CE; and (b) no sacrifices, so far as we know, were offered to Kongzi
outside of Kongzi’s hometown for the first two centuries of Western Han rule (206 BCE–8 CE).”
173
Han elite, they are a crucial part of the story. Once confirmed as an important text by Western
Han emperors (or, perhaps more accurately, by their handlers), the Lunyu was virtually guaranteed to circulate more broadly in elite society. Those who were most sensitive to the intellectual
and literary fashions at court were the officials with the education to compose memorials and the
access to submit them to the throne. Not only do early histories preserve a far greater number of
memorials than edicts, ruler-directed texts also represent a far more diverse array of interests and
intellectual perspectives than ruler-produced texts, which tend to be relatively conservative in
their outlook. Memorials are also invaluable sources of Kongzi quotation practice insofar as they
can be dated with a relatively high degree of accuracy. Some can be tied to specific days or
months or years, others to specific reigns or reign periods. That is not to say that all extant
memorials are authentic sources of the figures and occasions they represent. As we will see below, serious doubts can be raised about some of the earliest Han memorials to quote Kongzi and
the Lunyu.
The “Guardian” (bao 保) mentioned by Emperor Zhao in his 82 BCE edict was Huo Guang
霍光 (d. 68 BCE), a powerful official selected by Emperor Wu on his deathbed to act as regent
for Emperor Zhao.371 In 74 BCE, Huo Guang himself mentioned the Lunyu in a memorial submitted during the succession crisis which followed Emperor Zhao’s death. Just twenty-seven
days after being crowned as emperor, Liu He 劉賀 (b. 92), a grandson of Emperor Wu, was deposed by Huo Guang and other top officials on the grounds that he had acted in a “depraved and
disorderly” (yin luan 淫亂) manner.372 In his place, Huo Guang nominated Liu Bingyi 劉病已, a
great-grandson of Emperor Wu who would become Emperor Xuan, in a memorial which extolled
371.See Loewe 2000, pp. 170–71 for this episode.
372.Hanshu 8.238.
174
his nominee’s virtues:
[Concerning] the great-grandson of Filial Emperor Wu, [Liu] Bingyi, there was an imperial
edict [ordering] that he be reared and cared for in the Lateral Courts. At present, he is in his
eighteenth year. His teachers have taught him the Shi, the Lunyu, and the Xiaojing. He has
held to moderation and economy in his conduct and he is kind and humane and loves others,
[so that] he is capable of serving as an heir to Filial Emperor Zhao, of worshipping and serving
the founder of the house and his successors, and of treating [the people of] the ten-thousand
families as his children.
孝武皇帝曾孫病已，有詔掖庭養視，至今年十八，師受詩、論語、孝經，操行節儉，慈
仁愛人，可以嗣孝昭皇帝後，奉承祖宗，子萬姓。373
Here again we find an official document invoking the Lunyu in order to telegraph an imperial
candidate’s qualifications. The juxtaposition of Liu Bingyi’s educational program (the Shi, Lunyu, and Xiaojing) with his virtues (moderation, humaneness, aptitude for service) is no accident:
these texts were perceived as embodying the very qualities a virtuous ruler was expected to
exhibit.
In a memorial submitted to Emperor Cheng 成帝 (r. 33–7 BCE) in the early part of his reign,
the Chancellor (chengxiang 丞相) Kuang Heng 匡衡 (d. 30/29 BCE) defined the Lunyu’s value
in relation to the Classics as follows:
Your subject has heard that the Six Classics are the means by which the sages unified the
hearts of heaven and earth, manifested a return to morality, illuminated the lots of fortune and
misfortune, comprehended the rectification of the way of man, and prevented others from
going against their original nature. Thus when one examines the intent of the Six Arts, the
principles of men and heaven can be harmonized and the plants and animals can be cultivated.
This is the eternally changeless way. As for the Lunyu and Xiaojing, they are the essentials of
the sages’ words and conduct. One should investigate their meaning.
臣聞六經者，聖人所以統天地之心，著善惡之歸，明吉凶之分，通人道之正，使不悖於
其本性者也。故審六藝之指，則人天之理可得而和，草木昆蟲可得而育，此永永不易之
道也。及論語、孝經，聖人言行之要，宜究其意。374
Just as in the “Yiwen zhi,” in Kuang Heng’s memorial we find a division between canonical and
non-canonical texts, with the Lunyu and Xiaojing occupying some middle ground between them.
373.Hanshu 8.238. Translation adapted from Dubs 1955, vol 2, p. 204.
374.Hanshu 81.3343.
175
While Kuang Heng reserved his loftiest praise for the Six Classics, leaving his comments on the
Lunyu and Xiaojing almost as afterthoughts, he recommended studying them as sources of the
“essentials” (yao 要) of the sages’ teachings. It is tempting to imagine that the Lunyu’s appeal for
its elite audience was as a kind of shortcut text, something which promised direct access to
Kongzi’s teachings in a form that was both easily digestible and eminently quotable.
The perception of the Lunyu as an “essentials” text might have also motivated this passage
from a memorial submitted ca. 48 BCE by Jia Juanzhi 賈捐之 (fl. c. 48 BCE), the great-grandson of Jia Yi 賈誼 (c. 201–169 BCE):
Your subject has heard that Yao and Shun were the most illustrious sages and that Yu entered
the realm of the sages but did not surpass them. Thus Kongzi praised Yao as “truly great” [in
Lunyu 8/19], the Shao [music of Shun] as “thoroughly good” [in Lunyu 3/25], and Yu as
“faultless” [in Lunyu 8/21].
臣聞堯舜，聖之盛也，禹入聖域而不優，故孔子稱堯曰大哉，韶曰盡善，禹曰無間。375
The fact that these Lunyu quotations appear in three different chapters of the received Lunyu indicates that, like Liu Xiang and Wang Chong, Jia Juanzhi was thinking of the Lunyu as a single,
internally consistent text. More significantly, Jia Juanzhi looked to the Lunyu Kongzi to pronounce on Yao and Shun rather than quoting Yao and Shun speeches from the Shangshu. Perhaps the Lunyu Kongzi’s evaluations were the kinds of “essential” ideas praised by Kuang Heng
in his memorial, quotations which captured the core virtues of Yao and Shun more pithily than
the Classics themselves.
If we step back to consider the role of Kongzi and the Lunyu within the corpus of extant Han
memorials, we find additional evidence for a shift in Kongzi quotation practice in the middle part
of the Western Han. The following table presents the results of my survey of Western Han, Xin,
and Eastern Han memorials as collected by Yan Kejun in the Quan Han wen 全漢文 (Complete
375.Hanshu 64b.2831.
176
Texts of the [Former] Han) and Quan Hou Han wen 全後漢文 (Complete Texts of the Later
Han):376
total
Kongzi
attributions
Quan Hou Han 57
wen memorials
Kongzi attribu- explicit Lunyu
tions that paral- attributions
lel the Lunyu
37 (65%)
2
# of memorialists who cited
Kongzi
37
# of memorials
that cite
Kongzi
45
Quan Han wen
memorials
53 (80%)
23
30
66
5
In the aggregate, there is little to distinguish Western Han and Eastern Han memorials with respect to Kongzi and Lunyu quotation. But a closer look at the chronological distribution of
memorials with Lunyu-related Kongzi attributions reveals a familiar pattern (see Appendix 2:J).
Above we saw that Emperor Zhao was the first emperor to mention the Lunyu in an edict and
that Emperor Yuan was the first emperor to attribute a Lunyu saying to Kongzi. Most Kongzi and
Lunyu attributions in Han memorials likewise post-date Emperor Zhao’s reign.
Three documents buck this trend: a memorial supposedly submitted by Jia Yi 賈誼 (c. 201–c.
169) sometime between 174 and 169, Dong Zhongshu’s 董仲舒 (c. 179–c. 104) responses to a
series of questions posed by Emperor Wu in ca. 134, and a memorial submitted between 124 and
121 by Yuqiu Shouwang 吾丘壽王, who served as Grand Counsellor of the Palace (Taizhong
dafu 太中大夫) during Emperor Wu’s reign. The Jia Yi memorial credits Kongzi with a wordfor-word parallel of Lunyu 12/13 (“Kongzi said, ‘In hearing suits I am like other men, but I always make it so that there is no suit’” 孔子曰：聽訟，吾猶人也，必也使毋訟乎！) and the
Yuqiu Shouwang memorial quotes a close parallel of Lunyu 9/2 (“Kongzi said, ‘What should I
376.The Quan Han wen compilation also includes documents from the Xin 新 dynasty (9–23 CE). In addition to
“memorials” (zou 奏), my survey includes any and all texts labeled by Yan Kejun as having been “sent up” (shang
上) to the emperor. See Zhang Zuwei 2008 for a parallel effort to tabulate Lunyu quotations in the three histories.
177
take up? Should I take up archery?’” 孔子曰：吾何執？執射乎？).377 Dong Zhongshu’s responses to Emperor Wu incorporate one explicit Lunyu attribution and sixteen Lunyu Kongzi
sayings in what is the most spectacular use of Kongzi sayings in a Han memorial, and perhaps
even in any Han text.378 On the face of it, these documents would seem to indicate that the Lunyu
circulated during the reigns of Emperors Jing and Wu.
However, the authenticity of the Jia Yi and Dong Zhongshu memorials cannot be taken for
granted. Neither of these documents can be found in the Shiji, whose compilers Sima Tan (d. 110
BCE) and Sima Qian (d. c. 86 BCE) were contemporaries or near-contemporaries of Jia Yi and
Dong Zhongshu and thus might be expected to have been familiar with their writings. The earliest source for the memorials—indeed, the earliest source for all Western Han memorials that
quote Kongzi—is Ban Gu’s 班固 (32–92 CE) Hanshu, which was compiled well into the Eastern
Han period. The gap between the compilation of the Hanshu in the Eastern Han and the early
Western Han context raises the very real possibility that these documents did not originate in the
period they purport to represent. Indeed, Michael Loewe has noted that Dong Zhongshu’s second
memorial lists Kangju 康居 (Bactria) and Yelang 夜郎 (modern-day Guizhou 貴州 province)
among the regions which had submitted to the Han imperium even though these lands were not
incorporated into the Han empire until the Eastern Han period.379 This and other considerations
have led Loewe to conclude that “Chapter 56 of the Hanshu [i.e., Dong Zhongshu’s biography]
377.For Jia Yi’s memorial, see Hanshu 48.2253; for Jia Juanzhi’s memorial, see Hanshu 64a.2796.
378.The first memorial is at Hanshu 56.2498–2505; the second is at 56.2508–2513; the third is at 56.2514–2523.
The Lunyu Kongzi sayings in the first memorial are (in order) Lunyu 15/29, 7/14, 4/25, 12/19, 19/25, 20/2, 9/9, and
5/10; in the second memorial 13/12, 3/25, 3/25, 7/36, and 2/3; and in the third memorial 20/3, 15/5, 2/23. The third
memorial also includes a Kongzi yue saying which appears in the Xiaojing as a zi yue attribution: “Kongzi said, ‘Of
all the natures between heaven and earth, man’s is noblest” (孔子曰：“天地之性人為貴). See p. 100 for an excerpt
from one of these documents.
379.Loewe 2001, pp. 77–80. For the mentions of Bactria and Guizhou, see Hanshu 56.2525.
178
does not necessarily include Dong Zhongshu’s ipsissima verba, but what Ban Gu saw as being
his opinions and which suited the intellectual climate of the time and Ban Gu’s own purposes....Dong Zhongshu, a Chinese Pericles, may not himself have been as impressive a writer as his
biographer, a Thucydides.”380
Questions about authenticity aside, the sole Lunyu attribution in Dong Zhongshu’s memorials
is probably a later interpolation.381 As Kaneto Mamoru observed, the quotation in question—
“Your subject has heard that Lunyu [19/12] says, ‘Only a sage has both a beginning and an end’”
(臣聞論語曰：有始有卒者，其唯聖人虖！)—breaks the quotation pattern of the three memorials, which prefer Kongzi yue in every instance except this one.382 It also happens to be the only
example I have found in any early text in which a saying is prefaced by the mixed quotation
marker “Your subject has heard that the Lunyu says” (chen wen Lunyu yue 臣聞論語曰). Chen
wen was itself a common quotation marker which did not require additional clarification. My
guess is that the words “Lunyu” or “the Lunyu says” were added by some later editor who sought
to clarify the source of what was to him an obvious Lunyu quotation.
I suspect that Yuqiu Shouwang’s relative obscurity made him an unlikely target for later admirers/forgers. But interest in Jia Yi in the Han period was driven by his close association with
Qu Yuan 屈原, the tragic hero and supposed author of the Li sao 離騷 (Encountering Sorrow)
and “Huai sha 懷沙” (“Embracing sand”), two works in the Chuci 楚辭 (Verses of Chu) collec-
380.Loewe 2011, p. 121. However, Loewe also notes (2011, p. 87) that the Hanshu elsewhere (66.2896) preserves a
quotation attributed to “Master Dong” (Dong sheng 董生) which appears in the third of Dong Zhongshu’s
memorials (56.2521). Loewe also ultimately concludes (p. 336) that “[t]he three essays that Dong Zhongshu
presented in response to imperial rescripts are probably to be taken as some of the most authentic expressions of his
views,” despite the “signs of editorial work that was presumably accomplished by Ban Gu.”
381.Kaneto 1978, pp. 18–21. See also Makeham 1996, p. 12.
382.Hanshu 56.2514.
179
tion. According to his Shiji biography, Jia Yi also wrote a “Diao Qu Yuan fu 弔屈原賦”
(“Lament for Qu Yuan”) after suffering the slander of jealous officials and being dismissed from
the imperial court to serve a cruel regional king in Qu Yuan’s homeland.383 Like the Li sao, the
“Diao Qu Yuan fu” was read as a record of a virtuous official’s frustration at being unable to win
the favor and trust of his lord. In this way, Jia Yi was elevated as a latter-day Qu Yuan in the
Han imagination, and that reputation probably explains Ban Gu’s statement in his Hanshu evaluation (zan 贊) of Jia Yi that Emperor Wen 文帝, another figure retrospectively elevated in the
Han imagination as a counterweight to the perceived excesses of Emperor Wu’s reign, had devoted himself to virtue thanks to Jia Yi’s influence.384
Somewhat suspiciously, the Jia Yi memorial in question appears only in Jia Yi’s Hanshu biography, which closely tracks his Shiji biography except that it inserts an “abridged” (da lüe 大
略) version of the memorial in question which it introduces as follows: Jia Yi “frequently submitted memorials to explain government affairs because there were many things he wished to
correct and establish” (數上疏陳政事，多所欲匡建).385 The failure to mention a specific occasion or policy question is itself suspicious: extant memorials tend to have more clearly defined
contexts. But even if we take Jia Yi’s memorial at face value, a single Lunyu Kongzi saying
would be insufficient proof of the Lunyu’s existence in Jia Yi’s time given the possibility that
both Jia Yi and the Lunyu compilers might have drawn from a shared tradition.
Given these doubts, Yuqiu Shouwang’s memorial from the late 120s BCE that attributes an
383.Shiji 84.2491.
384.Hanshu 48.2265.
385.Hanshu 48.2230. Doubts have also been raised about the trustworthiness of Jia Yi’s Shiji biography. Shiji book
83, which includes the biographies of Qu Yuan 屈原 and Jia Yi, is one of several chapters to use the character tan
談, Sima Qian’s father’s name and a taboo character observed throughout the bulk of the Shiji. For this observation
see Bodde 1938, pp. 101–111. See also Kern 2003, pp. 306–07.
180
abbreviated version of Lunyu 9/2 to Kongzi should perhaps stand as the earliest memorial to reflect the possible influence of the Lunyu Kongzi.
• Histories
The edicts and memorials discussed above are invaluable sources insofar as they demonstrate the
Lunyu’s reach among the Han political elite. But memorials’ importance as sources of early Chinese history pales in comparison to the compilations in which most of these documents appear,
Sima Qian’s Shiji (Records of the Grand Historian), Ban Gu’s Hanshu (History of the Han), and
Fan Ye’s 范曄 (398–445) Hou Hanshu 後漢書 (History of the Later Han). These three histories
together supply so much of our knowledge about the early period that any biases in their treatment of Kongzi and Kongzi sayings become magnified in modern studies of the early period.
What do these texts, and the Shiji and Hanshu in particular, tell us about the evolution of Kongzi
quotation practice in the Han period?
Kongzi
attributions
Hou Hanshu (686,000 characters) 115
Kongzi attributions
with Lunyu parallels
78 (68%)
Lunyu
attributions
2
Hanshu (808,000 characters)
174
146 (84%)
8
Shiji (588,000 characters)
231
152 (66%)
0
At first glance, it would appear that the compilers of the three histories were all familiar with the
received Lunyu or something like it, with the Hanshu privileging the Lunyu slightly more than
the others. But these numbers mean relatively little in the aggregate. Significant differences between the Shiji’s and Hanshu’s use of Kongzi sayings emerge when we compare the internal distribution of Kongzi sayings and Lunyu parallels.
The Hanshu looks much like a typical Eastern Han text from the perspective of its Kongzi
181
quotations, which appear in fifty-five of the one-hundred chapters of the Hanshu. All but one of
the ten monographs (zhi 志) and half of the biographies (liezhuan 列傳) contain at least one
Kongzi saying with a Lunyu parallel, with the most Lunyu Kongzi sayings (19) appearing in
Hanshu 56, the biography of Dong Zhongshu. Hanshu Kongzi quotations also have parallels in
all twenty chapters of the received Lunyu without favoring any particular chapters (see Appendix
2:K). It would seem, then, that Ban Gu and the authors of his sources drew liberally from the
entirety of the received Lunyu or something closely resembling it.
In contrast, the distribution of Kongzi attributions in the Shiji is far less even. Two of the Shiji’s 130 chapters—chapter 47, the “Kongzi shijia 孔子世家” (“Hereditary House of Kongzi”),
and chapter 67, the “Zhongni dizi liezhuan 仲尼弟子列傳” (“Biographies of Kongzi’s Disciples”)—together account for 186 (81%) of Shiji Kongzi quotations and 124 (82%) of its Kongzi
sayings with Lunyu parallels. The other 128 chapters contain only 45 Kongzi attributions, or 19%
of the total, of which 28 (62%) have Lunyu parallels. The “Kongzi shijia” and “Zhongni dizi
liezhuan” excepted, Kongzi quotations have a much smaller presence in the Shiji than in the
Hanshu.
We can take this analysis one step further. Generally speaking, the Shiji might be said to consist of two kinds of text: the compiled material which Sima Qian (or Sima Tan 司馬談) drew
from in the course of arranging the Shiji, and the authored material which was composed specifically for the Shiji by Sima Qian. Of course, this distinction is not hard and fast, because Sima
Qian seems to have edited or even rewritten his sources as he saw fit, and he certainly did not
create new material ex nihilo. Generally speaking, however, determining when Sima Qian speaks
to us in his own voice and when he allows his sources to speak for themselves presents few difficulties. Examples of compiled material include edicts and memorials, treatises like Han Fei’s 韓
182
非 “Shui nan 說難” (“The Difficulties of Persuasion”) or Sima Tan’s 司馬談 discussion of the
liu jia 六家 (six schools of thought), excerpts from or summaries of canonical texts like the “Yao
dian 堯典” (“Canon of Yao”), and the numerous historical narratives woven together (one
presumes) from earlier chronicles and anecdotal traditions. Examples of authored material include Sima Qian’s “Zi xu 自序” (“Self-narration”) in which he explains the genesis of the Shiji,
the eight shu 書 (treatises) on topics ranging from ritual (Shiji 23) to currency (Shiji 30), the
chapter introductions which explain Sima Qian’s interest in his historical subjects, and the Taishi
gong yue 太史公曰 (The Grand Scribe says) comments which purport to provide Sima Qian’s
own thoughts on various topics and figures.386 In these sections of the text, Sima Qian speaks directly to his audience in order to shape their understanding of early history.
What proportion of Shiji Kongzi attributions are found in compiled versus authored materials? Remarkably, there are zero Kongzi attributions with Lunyu parallels in Shiji compiled material.387 All Lunyu and most non-Lunyu Kongzi sayings appear exclusively in Shiji authored
material.388
386.For example, the Kongzi saying discussed on p. 28 above (“The Master said, ‘My desire to record empty words
does not compare to my desire to see a person’s deeds in all their immediacy and vividness’”) appears in the Shiji
“Zi xu” and is thus an example of a Kongzi saying in authored material.
387.There are two possible exceptions. One is in a speech attributed to the Warring States-era persuader Cai Ze 蔡
澤 (Shiji 79.2422), which attributes a close parallel of Lunyu 7/16 to an anonymous “sage” (shengren 聖人) who
might have been Kongzi. However, since the speech does not mention Kongzi, it does not meet my minimum
qualifications for a Kongzi saying. The other exception is a Kongzi dialogue from the “Yue shu 樂書” (“Treatise on
music”) (Shiji 24.1226–30). Although the voice of the compiler dominates that chapter, the anecdote in question
seems to have been included as a self-contained text and thus could be considered “compiled” material.
388.By my count there are thirteen Kongzi sayings with Lunyu parallels in Taishigong yue comments; four in the
treatises (shu 書); two in chapter introductions; one in an evaluation of a historical figure; and five in Shiji book 61,
the biography of Boyi 伯夷 and Shuqi 叔齊, the very first Shiji biography and the text which explains Sima Qian’s
interest in biography generally. There are also a handful of Kongzi sayings without Lunyu parallels in Shiji compiled
material, but in all of those examples the Kongzi saying is appended to a narrative and functions like a Zuozhuan
Kongzi comment insofar as it steps out of the narrative to evaluate the historical figure in question. One of these
sayings (Shiji 39.1675) also appears at Zuozhuan Xuan 2 (SBCK 10/4a; KZJY p. 414): “Kongzi learned of this and
said, ‘Dong Hu is a good scribe in the ancient mould. In writing [in accordance with] the law he hid nothing. Xuanzi
is a fine grandee, and for the sake of the law he accepted blame. How unfortunate! He would have escaped
[punishment] had he left the borders [of his state]’” (孔子聞之，曰：董狐，古之良史也，書法不隱。宣子，良
183
This is an important result with major implications both for the study of Kongzi and for early
Chinese history generally. The Shiji Kongzi is the historiographer’s Kongzi, an authority appealed to in order to make sense of and evaluate the various figures and events in the text. Shiji
Kongzi sayings were overlaid on top of early Western Han and pre-Han sources; by and large,
they were not part of the sources themselves. Consequently, there is little reason to assume that
Sima Qian’s Kongzi would have been familiar to the authors of Sima Qian’s sources. Perhaps we
even overestimate the importance and visibility of the Lunyu Kongzi in the pre-Han context because of his prominence in Shiji authored material. Moreover, the use of Kongzi sayings primarily within authored material, presumably the last layer to have been added to the Shiji, is consistent with the theory that the Lunyu did not begin to circulate until some point during Sima Qian’s
lifetime. If the Lunyu circulated prior to Sima Qian’s lifetime, we might expect to find Lunyu
Kongzi sayings quoted more widely in Sima Qian’s sources.
What was the precise relationship between the Shiji and Lunyu? As noted above, Lunyu parallel sayings account for roughly 66% of all Kongzi attributions in the Shiji with the remaining
44% deriving principally from the Zuozhuan and other unknown sources. Considered on its own,
the 66% figure is prima facie evidence for treating Lunyu parallel sayings as actual Lunyu quotations, a conclusion encouraged by the Taishigong yue comment to the “Zhongni dizi liezhuan”:
His Excellency the Grand Scribe said, “Scholars often refer to the seventy disciples. Those
who praise them sometimes go beyond the facts, and those who criticize them sometimes miss
the truth of the matter. When I weigh [these accounts, I see that such scholars] assessed [the
disciples’] sayings without having observed their comportment.389 The disciple records came
out of the ancient writings of the Kong clan—they are close to being correct. Using these I
took all of the disciples’ various names from the disciple questions in the Lunyu and arranged
them together into a single chapter, while omitting the suspect passages.
太史公曰：學者多稱七十子之徒，譽者或過其實，毀者或損其真，鈞之未睹厥容貌則論
大夫也，為法受惡。惜也，出疆乃免。).
389.This point echoes the Kongzi saying from Shiji book 130 discussed in chapter one (p. 28).
184
言。弟子籍出孔氏古文，近是。余以弟子名姓文字悉取論語弟子問并次為篇，疑者闕
焉。390
However, Kaneto Mamoru 金戶守 has argued that this comment cannot be taken at face value
because the Lunyu could not have been the primary source for the “Zhongni dizi liezhuan.”391
The chapter names seventy-seven of Kongzi’s disciples with each entry including information
about a disciple’s names, birthplace, and age in relation to Kongzi. Consider the entry for Zeng
Shen 曾參, aka Zengzi 曾子, the purported author of the Xiaojing and one of the most prominent
disciples in the Lunyu:
Zeng Shen was a man of southern Wucheng and his style name was Ziyu. He was younger
than Kongzi by 46 years. Kongzi believed that he was capable of mastering the Way of filial
piety and so imparted his teachings to him. He authored the Xiaojing and he died in Lu.
曾參，南武城人，字子輿。少孔子四十六歲。孔子以為能通孝道，故授之業。作孝經。
死於魯。392
Curiously, the Lunyu does not include any information about Zengzi’s birthplace, style name (zi
字), age, authorship of the the Xiaojing, or death,393 and this problem is by no means exclusive to
Zengzi’s biography. The received Lunyu mentions only thirty of the seventy-seven disciples in
the “Zhongni dizi liezhuan,” and for those thirty it supplies little to no information about their
names, birthplaces, etc. and no information whatsoever about their ages. Assuming that Sima
Qian’s Lunyu resembled our own, it simply cannot be the case that Sima Qian “took all the disci-
390.Shiji 67.2226. Nienhauser 1995 understands dizi ji 弟子籍 as the title of a text called “The Register of
Disciples.” However, I see no need to read dizi ji as a named text instead of as a generic label for a collection of
writings, nor is there any evidence to suggest that such a text existed prior to the Shiji’s compilation. On the other
hand, book 38 of the Kongzi jiayu, “Qishier dizi jie 七十二弟子解” (“Explicating the Seventy-two Diciples”),
consists of a number of Kongzi disciple biographies and might conceivably reflect a pre-Shiji tradition of Kongzi
disciple records.
391.Kaneto 1974, pp. 17–19. Here we might recall Huan Tan’s 桓譚 (43 BCE–28 CE) claim, preserved in a
fragment collected by Yan Kejun in the Quan Hou Han wen (15/4b; p. 549), that the Taishi gong yue comments in
the Shiji were authored by Dongfang Shuo 東方朔.
392.Shiji 67.2205
393.However, Lunyu 8/3 and 8/4 open with the lines “Zengzi had a serious illness” (曾子有疾).
185
ples’ various names from the Lunyu disciple questions.”
More problematic still is the fact that not one of the fourteen sayings attributed to Zengzi in
the Lunyu appears in the “Zhongni dizi liezhuan” or anywhere else in the Shiji,394 despite
Zengzi’s prominence in the received Lunyu.395 As we have seen (p. 91), Lunyu 4/15 famously
credits Zengzi with having identified the single idea which runs through Kongzi’s teachings. But
the sole Zengzi-related Lunyu passage with a parallel in the Shiji is Lunyu 11/18, which criticizes
a certain Shen 參, presumably Zeng Shen or Zengzi, as being lu 魯 (obtuse).396
The complete absence of Zengzi sayings in the Shiji supports Kaneto’s argument that the received version of the Lunyu postdates the Shiji and, in fact, drew from the Shiji for its Kongzi
and Kongzi disciple sayings. This view commits Kaneto to arguing that the Lunyu mention in the
Taishigong yue comment to the “Zhongni dizi liezhuan” is a later interpolation, one that contradicts Sima Qian’s statement that the “disciple records [which] came out of the ancient writings of
the Kong clan” (弟子籍出孔氏古文) were his primary sources for the chapter. Kaneto’s theory
aside, if we assume that Sima Qian had access to a Lunyu text, the lack of Lunyu Zengzi parallels
plus the distribution of Shiji Kongzi parallels in the Lunyu might indicate that Sima Qian’s Lunyu
looked significantly different from our own. The Shiji includes no parallels from Lunyu 16, 19,
and 20 and only one parallel from Lunyu 8 (see Appendix 2:L); perhaps these chapters, like Lunyu Zengzi material, were added to the Lunyu at some point after the compilation of the “Kongzi
394.There is one possible exception, an unattributed parallel to Lunyu 8/4 which appears at Shiji 126.3208.
However, that quotation is prefaced by “a tradition says” (zhuan yue 傳曰) and not “Zengzi says.”
395.These are Lunyu 1/4, 1/9, 4/15, 8/3, 8/4, 8/5, 8/6, 8/7, 12/24, 14/26, 19/16, 19/17, 19/18, 19/19.
396.Shiji 67.2185. In addition, an anecdote from Mengzi 3A/4 featuring Zixia, Zizhang, Ziyou, You Ruo, and
Zengzi appears at Shiji 67.2216 but without any mention of Zengzi. In the Mengzi version of the anecdote, Zengzi
criticizes the other disciples for setting up You Ruo 有若 (i.e., “He who has a resemblance”) as their master after
Kongzi’s death.
186
shijia” and “Zhongni dizi liezhuan.”
Unfortunately, Kaneto was unaware of the Dingzhou manuscript find when he developed his
theory of a post-Shiji Lunyu, otherwise he might have been more willing to consider the possibility that the Shiji drew from a Lunyu text. If a text closely resembling the received Lunyu was already circulating among the Han elite by 55 BCE, it is not hard to imagine Sima Qian having access to a Lunyu text a few decades earlier. That problem aside, it is worth noting that the Shiji is
the first text encountered in this survey about which it has been argued that the Lunyu compilers
drew from it rather than the other way around.
Interlude: Searching for signs of the Lunyu
Thus far I have discussed various signs of the Lunyu’s presence, influence, and authority in the
early period. These include:
● Explicit mentions. A text called “Lunyu” is mentioned throughout Eastern Han texts and to a
lesser extent in Western Han texts. Emperors listed it as one of the texts they had studied, officials encouraged emperors to see it as a record of “the essentials of the sages’ words and deeds,”
and bibliographers ranked it alongside the Six Classics.
● Explicit attributions. Numerous Han authors attributed Kongzi sayings (and Kongzi disciple
sayings) to the Lunyu itself. For some Eastern Han commentators, it seems to have mattered less
that a given Lunyu quotation originated with Kongzi than that it appeared in a widely circulating,
and thus widely recognizable, text like the Lunyu.
● High fidelity to the Lunyu versions of Kongzi sayings. Not only did the Han authors considered
187
thus far quote extensively from the Lunyu, they also strayed very little from the Lunyu versions
of Kongzi sayings. Most of the examples considered above are word-for-word Lunyu quotations.
One exception that proves the rule is Wang Mang’s quotation of Lunyu 13/3 in a 20 CE edict
which abbreviated the Kongzi saying in a way that acknowledged the abbreviation.
● Reliance on the Lunyu as the primary source of Kongzi sayings. Even if Han texts had not preserved a single mention of a “Lunyu” text, one could infer its presence from the ratio of Kongzi
sayings with Lunyu parallels to those without. By my count, Lunyu parallel sayings outnumber
non-Lunyu Kongzi sayings by a ratio of approximately 3½:1 in the Eastern Han. (That figure
jumps to more than 6½:1 if one includes explicit Lunyu yue quotations together with Kongzi yue
quotations.) The predominance of Lunyu Kongzi sayings in the texts considered here is the surest
indicator that authors in the period did not just pay lip service to the Lunyu. They also used the
text as if it was most authoritative source of Kongzi sayings.
● Use of the entire Lunyu. Earlier I noted that the Hanshu took its Lunyu quotations from all
twenty chapters of the received Lunyu without privileging any single chapter or set of chapters.
Such consistent quotation suggests that Ban Gu’s Lunyu looked much like our own.
● Reliance on the Lunyu as a point of “reference” (an 案/按) for suspect textual traditions. This
was how Wang Chong used the text, pointing out that certain Kongzi sayings or anecdotes were
spurious because they did not appear in the Lunyu (or in the Five Classics).
● Use as proof text. Most of the Kongzi attributions considered thus far, and all Kongzi quotations in Han edicts and memorials, were meant to buttress the arguments of the quoting authors.
Lunyu Kongzi sayings would not have suited this purpose so well if the Lunyu was not viewed as
an authoritative source.
● Lunyu commentaries. The commentaries preserved in the Lunyu jijie and listed in the “Yiwen
188
zhi” are clear proofs of the Lunyu’s importance in the period. In the Shuiyuan, a text with a relatively high ratio of non-Lunyu Kongzi sayings to Lunyu Kongzi sayings, Liu Xiang’s interest in
explaining Lunyu Kongzi sayings indicates the Lunyu’s authority relative to other sources of
Kongzi sayings.
● Inspiration for new texts. Authors like Yang Xiong and Zhao Qi sold their texts as “imitations”
(xiang 象) of the Lunyu. Texts like Wang Fu’s Qianfu lun also included first chapters on the subject of xue 學 (learning) in apparent imitation of Lunyu book 1. Such imitators established the
Lunyu as a text worthy of emulation.
● Pedagogical and symbolic value. The Lunyu was mentioned alongside the Shi, Shu, and Xiaojing as one of the few texts a crown prince or underage emperor was expected to have studied.
The Lunyu could also be invoked as a symbol of studiousness, scholarly aptitude, and virtue. In
Ban Gu’s account of Wang Mang’s demise, his adaptation of a Lunyu Kongzi saying symbolized
his perversion of canonical texts more generally.
This analysis of Kongzi quotation practice in the latter part of the early period will focus our survey as we move further back in time. Can any of these indices of the Lunyu’s authority, influence, or circulation be found in early Western Han and pre-Han texts? My answer to that question is (a qualified) no. Not only do early Western Han and pre-Han texts preserve no explicit
Lunyu mentions or attributions, they also provide very few if any indications that Lunyu Kongzi
sayings were treated differently compared to non-Lunyu Kongzi sayings.
The early Western Han period (203–100 BCE)
189
From this point backwards, the dating of so many texts traditionally dated to the early Western
Han and pre-Han periods becomes harder and harder to verify. One exception is the Huainanzi
淮南子, a text in twenty-one chapters and approximately 131,000 characters which was compiled under the auspices of Liu An 劉安 (179–122 BCE), the King of Huainan 淮南, and presented to the newly enthroned Emperor Wu in 139 BCE.397 As noted earlier, compared to the Shiji and later compilations of comparable scale, the Huainanzi exhibits a relatively small
proportion of Kongzi sayings with Lunyu parallels. Although somewhat larger than Liu Xiang’s
Shuiyuan, the Huainanzi contains only twenty-one Kongzi sayings, or roughly 10% of the
Shuiyuan’s total. Only three of these sayings (14%) parallel the Lunyu. Nowhere does the
Huainanzi explicitly mention a Lunyu text.
Two of the Huainanzi’s three Kongzi sayings are word-for-word parallels of Lunyu 13/6 and
9/30 (see Appendix 2:M, parallels 1 & 2). The first saying corroborates the idea that a ruler need
only rectify himself in order to be obeyed by his people, and the second introduces a Kongzi yue
proof text which is then elaborated much as Liu Xiang elaborated Lunyu Kongzi sayings in the
Shuiyuan. While their use as proof texts hint at the Huainanzi compilers’ awareness of, and respect for, a Lunyu text, there is a crucial difference between the Huainanzi’s handling of Lunyu
Kongzi sayings compared to later texts’: the Huainanzi does not reserve such treatment for Lunyu Kongzi sayings alone. Consider the following quotations from Huainanzi books 10 and 11:
<Huainanzi book 10>Learning of the good is easy, using it to correct oneself is difficult.
When the Master saw the three transformations of grain [from seed to sprout to mature plant]
he let out a great sigh and said, “The fox dies facing his den, but my head [droops like] grain.”
Thus when the noble man sees the good he suffers in himself. If he corrects himself embracing
the distant is easy. Thus a Shi [#191] says, “One who is not personally involved or affectionate
will not be trusted by the people.”
397.For an introduction to the Huainanzi, see Major et al. 2010, pp. 1–13. Vankeerberghen 2001 is an excellent
treatment of the text’s political background.
190
聞善易，以正身難。夫子見禾之三變也，滔滔然曰：狐鄉丘而死，我其首禾乎！故君子
見善則痛其身焉。身苟正，懷遠易矣。故詩曰：弗躬弗親，庶民弗信。398
<Huainanzi book 11>Human nature is both the Great Dipper and the North Star for mankind.
When one has some way to see himself he does not lose [his grasp of] the underlying
dynamics of things; when one has no way to see himself he is confused and entrapped with
every movement...Kongzi said to Yan Hui, “I serve you by forgetting [me], and you serve me
by forgetting [me]. Nonetheless, even though you forget me there is still something
unforgotten which is preserved.” Kongzi understood the root.
夫性，亦人之鬥極也。有以自見也，則不失物之情；無以自見，則動而惑營。...孔子謂
顏回曰：“吾服汝也忘，而汝服於我也亦忘。雖然，汝雖忘乎吾，猶有不忘者存。孔子
知其本也。399
The juxtaposition of a non-Lunyu Kongzi saying with a Shi proof text in the first example is
telling. In the Shuiyuan and later texts only Lunyu Kongzi sayings functioned as proof texts. But
the author of Huainanzi book 10 treated this non-Lunyu Kongzi saying as a proof text no less authoritative than the Shi. In the second example, a non-Lunyu Kongzi saying is used both as a
proof text and as evidence of Kongzi “understanding the root” (zhi ben 知本). In contrast to
Wang Chong, the author of Huainanzi book 11 was apparently content to rely on non-Lunyu material to arrive at conclusions about Kongzi’s character.400
The third Kongzi saying with a Lunyu parallel, a variant of Lunyu 2/19 from Huainanzi book
16, differs so much from its Lunyu counterpart as to make it an unlikely Lunyu quotation (see
Appendix 2:M, parallel 3). Despite their shared interest in distinguishing “raising the crooked”
(ju wang 舉枉) from “raising the straight” (ju zhi 舉直), it is the differences between these parallels that stand out. Whereas the Lunyu version casts the saying as advice to Duke Ai, the
Huainanzi version has no interlocutor and instead treats it as Kongzi’s defense of his involvement with the Jisuns 季孫, one of three clans who wrested effective control of Lu away from the
398.Huainanzi 10 (SBCK 10/4b; KZJY 3/29, p. 37). Translation adapted from Major et al. 2010, p. 361.
399.Huainanzi 11 (SBCK 11/4a; KZJY 17/12, p. 391). Translation adapted from Major et al. 2010, p. 404.
400. For another example, see the Kongzi saying discussed in the conclusion to chapter one (p. 130).
191
royal family. I shall have more to say about the relationship between these two Kongzi sayings in
chapter three (p. 233). In any event, the case for seeing the Huainanzi as a post-Lunyu text would
seem to rest solely on the parallels to Lunyu 9/30 and 13/6, which together account for only thirty
of the roughly 16,000 characters in the received Lunyu (0.2%).
The Huainanzi is by no means an outlier in the early Western Han period. The Mawangdui
馬王堆 corpus, a collection of manuscripts excavated in the early 1970s from a tomb dated to
168 BCE, contains over one hundred Kongzi yue and zi yue sayings within a number of texts related to the Yi 易. None of these sayings parallel the Lunyu. Likewise, the list of Kongzi-related
anecdote titles written on a wooden board discovered in the Fuyang 阜陽 find, a tomb closed in
165 BCE, contains not a single entry with an obvious Lunyu parallel. A number of other texts
thought to date to the Western Han period, including the Guliang 穀梁 commentary to the Chunqiu, the Xiaojing 孝經 (Classic of Filial Piety), and Jia Yi’s 賈誼 Xinshu 新書 (New Writings),
contain no Lunyu Kongzi sayings or Lunyu mentions.401 In this period one struggles to find evidence of the Lunyu’s existence, let alone its authority.402
Two important compilations with possible origins in the early to mid-Western Han complicate this conclusion. These are the Liji 禮記 (Ritual Records) and the Hanshi waizhuan 韓詩外
傳 (Outward Commentary on the Han Odes). Although the pre-Han origins of a few Liji chapters
have been confirmed by recent manuscript finds and/or the presence of substantial parallels in
other pre-Han texts,403 determining the chronology of most of the 45 chapters of the Liji presents
401.The Guliang and Xinshu have no Lunyu Kongzi sayings. For the Xiaojing’s Lunyu 17/21 parallel, see SBCK
1/13b.
402.See also Takeuchi 1972, p. 80 for the observation that early W Han texts sometimes parallel the Lunyu,
sometimes not.
403.Of particular importance are the Guodian and Shanghai Museum manuscript versions of “Zi yi 緇衣” (“Black
Robes”) and the Shanghai Museum version of “Kongzi xianju 孔子閒居” (“Kongzi resting in leisure”). For the
Guodian “Zi yi,” see Guodian Chu mu zhujian pp. 126–137; for the Shanghai Museum “Zi yi” and “Min zhi fumu,”
192
tremendous difficulties.404 The Hanshi waizhuan is traditionally attributed to Han Ying 韓嬰, an
erudite (boshi 博士) during the reign of Emperor Wen 文帝 (r. 180–157 BCE) and Grand Tutor
(Taifu 太傅) to a regional king during the reign of Emperor Jing 景帝 (r. 157–141 BCE).405 One
of the most widely paralleled texts from the period, the Hanshi waizhuan shares a significant
amount of material with the Xunzi, Liji, Da Dai liji, Shuiyuan, Xinxu, and Lienü zhuan, among
other texts.406 The high degree of intertextuality creates an impression of a collection cobbled together from disparate sources, quite possibly at different stages throughout the Western Han or
later.407
Kongzi sayings with Lunyu parallels account for only 30 (9%) of the 320 Kongzi attributions
in the Liji and 14 (13%) of the 112 Kongzi attributions in the Hanshi waizhuan, well below the
average in the Shiji and later texts. By my count only three of the 28 Lunyu parallel sayings in
the Liji do not exhibit significant variants,408 and the Liji does not appear to privilege Lunyu
Kongzi sayings over non-Lunyu sayings. On the other hand, the majority of Lunyu Kongzi say-
see Shanghai bowuguan cang Zhanguo Chu zhushu vol. 1, pp. 169–200, and vol. 2, pp. 151–180. Shanghai
bowuguan cang Zhanguo Chu zhushu vol. 1, pp. 201–213, contains a comparison of the Guodian and Shanghai
Museum “Zi yi” manuscripts. The Liji chapter “Yue ling 月令” (“Monthly ordinances”) overlaps with the Lüshi
chunqiu, and “San nian wen 三年問” (“Asking about the three-year [mourning period]” and “Yue ji 樂記” (“Record
of music”) have substantial parallels in the Xunzi.
404.On the dating of the Liji, see Loewe (ed.) 1993, pp. 293–297, although Jeffrey Riegel’s entry there does not
include information on recent manuscript finds.
405.For a summary of what is known about Han Ying, see Loewe 2000, p. 151.
406.Hightower 1952, p. 3.
407.Riegel (1986, p. 20) also questions the traditional dating of the Hanshi waizhuan as well as the “Fang ji.”
Contrast this approach with that of Haupt (2006, pp. 104–107) and Makeham (1996, pp. 12–13), who take for
granted that the Hanshi waizhuan can be dated to Han Ying’s lifetime.
408.For these three sayings, see the Lunyu 17/21 parallel in “San nian wen 三年問” (“Questions about the threeyear mourning period”) (SBCK 18/12b), the Lunyu 12/13 parallel in “Da xue 大學” (“The Great Learning”) (SBCK
19/9b), the Lunyu 3/7 parallel in “She yi 射義” (“The principles of archery”) (SBCK 20/11a), and the Lunyu 6/29
parallel in “Zhong yong 中庸” (“Doctrine of the Mean”) (SBCK 16/1b). This list does not include a number of
parallels to Lunyu book 10, which are not prefaced by zi yue 子曰. For a list of book 10 parallels, see Appendix 2:V
and Weingarten 2009.
193
ings in the Hanshi waizhuan are fairly close parallels of their Lunyu counterparts,409 and the handling of some of these sayings indicates that they were considered relatively authoritative. Consider this Lunyu 20/3 parallel from Hanshi waizhuan book 6:
The Master said, “Without knowing the mandate one has no way of being a noble man.” This
says that everything heaven gives life to has a heart of humaneness, rightness, ritual,
knowledge, conformity, and goodness. If you do not know how heaven commands life then
you do not have a heart of humaneness, rightness, ritual, knowledge, conformity, and
goodness. One who does not have a heart of humaneness, rightness, ritual, knowledge,
conformity, and goodness is called a “petty man.” Thus it is said, “Without knowing the
mandate one has no way of being a noble man.”
子曰：不知命，無以為君子。言天之所生，皆有仁義禮智順善之心，不知天之所以命
生，則無仁義禮智順善之心，無仁義禮智順善之心，謂之小人。故曰：不知命，無以為
君子。410
It is probably no coincidence that the author of this passage, like Liu Xiang in his explanation of
Lunyu 6/2, chose to elaborate a Lunyu Kongzi saying; no non-Lunyu Kongzi sayings in the Hanshi waizhuan is featured in this way.411 But it also happens to break the only discernable organizational principle in the Hanshi waizhuan, a text in which nearly every section is capped with at
least one Shi quotation. The lack of Shi quotation in this instance might hint at some sort of textual corruption, in which case the Lunyu Kongzi saying might not reflect an early Western Han
context.
The four Lunyu yue attributions in the Liji and Hanshi waizhuan, one in the “Fang ji 坊記”
(“Embankment Record”) chapter of the Liji and one each in books two, five, and six of the Han-
409.However, Haupt (2007, p. 107) has observed that the Hanshi waizhuan Lunyu parallels consist only of short
phrases that appear in very different contexts. They also never include the entirety of a Lunyu entry. For instance, in
Hanshi waizhuan 2/16, the Lunyu parallel is confined to the last eleven characters of a 63-character saying and
follows an anecdote in which Zilu questions Kongzi on the propriety of his meeting with a certain Cheng Benzi of
Qi 齊程本子, a context with no parallel in the Lunyu.
410.Hanshi waizhuan 6 (SBCK 6/9a). Translation adapted from Hightower 1952, p. 206.
411.See Haupt 2006, pp. 101–107, for her discussion of the question of whether the Lunyu was a source for the
Hanshi waizhuan. Her conclusion (pp. 135–140) is that a proto-Lunyu collection of Kongzi sayings was probably
available to the Hanshi waizhuan compiler.
194
shi waizhuan,412 are among the candidates for the earliest mentions of a “Lunyu” text in the extant literature.413 However, the fact that the “Fang ji” Lunyu quotation is the only such quotation
in the whole of the Liji is reason enough to suspect its authenticity.414 The three Hanshi waizhuan
Lunyu attributions, on the other hand, cannot be dismissed so easily. Consider this passage from
Hanshi waizhuan book five:
Kongzi was sitting in attendance on the Jisuns when Prime Minister Tong said, “If the lord
sends someone to acquisition a horse, should it be given to him?” Kongzi said, “I have heard
that when a lord takes something from his subject he calls it an ‘acquisition,’ not a
‘borrowing.’” The Jisuns realized this and so instructed Prime Minister Tong, saying,
“Henceforth when the lord acquisitions something it will be called an ‘acquisition,’ not a
‘borrowing.’” When Kongzi rectified the words ‘borrowing a horse’ the principle of rulersubject [relations] was settled. The Lunyu said, “I would invariably rectify terminology.” A
Shi says, “A noble man does not make light of his words.”
孔子侍坐於季孫。季孫之宰通曰：君使人假馬，其與之乎？孔子曰：吾聞君取於臣，謂
之取，不曰假。季孫悟，告宰通曰：今以往，君有取，謂之取，無曰假。孔子曰正假馬
之言，而君臣之義定矣。論語曰：必也正名乎。詩曰：君子無易由言。415
This is the only example I have encountered in any early text in which a Lunyu quotation caps a
non-Lunyu Kongzi saying. While the substance of the Lunyu quotation is entirely consistent with
Kongzi’s advice to the Jisun clan, the statement that Kongzi “rectified the words ‘borrow a
horse’ and thereby settled the principle of lord-subject relations” (孔子曰正假馬之言，而君臣之
義定矣。) echoes Lunyu 13/3 so closely as to make the Lunyu quotation seem redundant in con-
text. This leads me to suspect that the Lunyu quotation might have been added by a later editor
who also noticed the parallel with Lunyu 13/3. Perhaps the same argument could be made of the
412.SBCK 15/15a.
413.Makeham 1996, p. 10–13.
414.It also bears mentioning that the “Fang ji” is a mess of a text. Although it contains hints of systematic
organization (e.g., the repetition of the phrase “use this as a dyke for the people” [以此坊民]), those hints are not
consistent and the logic of the chapter is far from obvious. My suspicion is that the Lunyu attribution crept into the
text after the earlier, more coherent organization was disrupted. See Kern 2005 for a description of a similar
phenomenon in the manuscript and received versions of the “Zi yi 緇衣” (“Black Robes”) chapter of the Liji.
415.Hanshi waizhuan 5 (SBCK 5/17b). Following Hightower (1952, p. 190), I have omitted the yue 曰 of Kongzi
yue zheng jia ma 孔子曰正假馬 in keeping with the Xinxu (SBCK 5/9a) version of this passage.
195
two other Lunyu attributions in the Hanshi waizhuan.416
In any event, my suspicion is that both the Liji and the Hanshi waizhuan consist of multiple
layers and multiple layers within layers, some of which were added by authors or editors who
were already familiar with a Lunyu text. This would make using them as sources for early Western Han or pre-Han Kongzi quotation practice extremely problematic.
The Pre-Han Period
• The Easy Cases
Signs of the Lunyu’s influence become even fainter as we move backward into the pre-Han period. Nine prominent texts typically dated to the pre-Han period contain fewer than five Kongzi
sayings with Lunyu parallels apiece (for these parallels, see Appendices 2:N–W):
Lüshi chunqiu 呂氏春秋 (c. 239 BCE)
Total Kongzi
attributions
34
Kongzi sayings with Lunyu
Lunyu parallels
attributions
4 (12%)
0
Gongyang zhuan 公羊傳 (3rd c. BCE?)
8
1 (12%)
0
Han Feizi 韓非子 (Han Fei 韓非, c. 280–c. 33
233 BCE?)
1 (3%)
0
Xunzi 荀子 (Xun Qing 荀卿, d. c. 250
BCE?)
38
2 (5%)
0
Zhouyi 周易 (3rd century?)
30
0 (0%)
0
Yanzi chunqiu 晏子春秋
15
1 (7%)
0
Zhuangzi 莊子 (Zhuang Zhou 莊周, c. 369– 92
286 BCE?)
3 (3%)
0
416.One of these is a quotation of Lunyu 10/27 in Hanshi waizhuan book 2 (SBCK 2/12b), the other is a quotation of
Lunyu 13/3 in Hanshi waizhuan book 6 (SBCK 6/4a). To my eye, both quotations seem just as tacked on as the
Lunyu 13/3 quotation, although I readily admit the subjective nature of such judgments.
196
Zuozhuan 左傳 (4th century?)
42417
Mozi 墨子 (Mozi 墨子, c. 470–391 BCE?) 3
3 (7%)
0
2 (66%)
0
Strikingly, only fifteen (6%) of the 250 Kongzi sayings in these nine texts have parallels in the
Lunyu, and only two of these—a Lunyu 11/23 parallel from the Lüshi chunqiu (appendix 2:N,
parallel 1) and a Lunyu 11/9 parallel from the Gongyang—are word-for-word Lunyu parallels.
The seventeen other Lunyu parallel sayings all exhibit significant variants when compared with
their Lunyu counterparts.
The scarcity of Lunyu Kongzi sayings in the Lüshi chunqiu and Zuozhuan is particularly
noteworthy. The Lüshi chunqiu is a large compilation (roughly 100,000 characters) whose authors aimed to produce a comprehensive summa of received wisdom as a guide for rulers. Even
if the Lüshi chunqiu authors were aware of a Lunyu text, they did not treat its Kongzi sayings as
especially authoritative. Consider the Lunyu 9/28 parallel from the “Shen ren 慎人” (“Being
mindful of others”) section of the Lüshi chunqiu (see Appendix 2:N, parallel 4). That parallel
(“When the extreme cold arrives...”) is but one line in an extended dialogue between Kongzi and
his disciples, and the dialogue gives no indication that this particular line was more important
than any other. I discuss this example at length in chapter three (p. 231).
A text of roughly 196,000 characters, the Zuozhuan is almost twice as large as the Lüshi
chunqiu but contains fewer Lunyu parallels among its Kongzi sayings. Perhaps the most striking
parallel is a saying which appears at Lunyu 12/1 and in the Zuozhuan entry for the twelfth year of
Duke Zhao’s 昭公 reign (see Appendix 2:T, parallel 1). Although both versions feature the same
core aphorism (“Overcoming oneself and returning to ritual is humaneness” 克己復禮，仁也),
417.In contrast, Haupt (2006, pp. 127–128) finds two of twenty-six Kongzi sayings in the Zuozhuan have parallels
in the Lunyu.
197
the Lunyu presents it as Kongzi’s words whereas the Zuozhuan Kongzi attributes it to “an ancient
maxim” (gu ye you zhi 古也有志). One presumes that the author of the Zuozhuan comment
would not have framed it in this way had he been quoting the Lunyu.
Another ostensibly pre-Han Lunyu parallel appears in Mozi book 46, which features Mozi
criticizing Kongzi for his advice to Zigao, the Duke of She 葉公子高, to “attract the distant and
renew the old” (遠者近之，而舊者新之。). This passage has been cited as evidence that the
Mozi was written by someone with knowledge of a Lunyu text, and specifically Lunyu 13/16:418
<Lunyu 13/16>The Duke of She asked about governance. The Master said, “Please those near
at hand and attract those far away.”
葉公問政。子曰：近者說，遠者來。
<Mozi book 46>Zigao, the Duke of She, asked Kongzi about governance, saying, “What about
successful governance?” Zhongni responded, “Successful governance entails attracting the
distant and renewing the old.” Mozi learned of this and said, “Zigao, the Duke of She, did not
understand what to ask, and Zhongni did not understand how to respond.”
葉公子高問政於仲尼曰：善為政者若之何？仲尼對曰：善為政者，遠者近之，
而舊者新之。子墨子聞之曰：葉公子高未得其問也，仲尼亦未得其所以對
也。419
However, as noted in chapter one (p. 96), the idea of “pleasing the near and attracting the distant” appears in a number of early texts, including the Xunzi 荀子, Guanzi 管子, Wenzi 文子,
and Hanshu 漢書, without any association with Kongzi (see Appendix 2:U). It is even attributed
to Mozi himself in Mozi books 10 and 35:
<Mozi book 10>Pacify those near at hand and attract those far away.
近者安之，遠者歸之。420
<Mozi book 35>Pacify the governance of those near at hand and attract those far away to your
418.Goldin 2001, pp. 94–95. Goldin notes that “[m]ost scholars consider the ‘Mohist Analects’ [book 46] a reliable
collection of material dating from the turn of the fifth century B.C.E. and earlier; indeed, it is by far the least
problematic part of the entire Mozi.”
419.Mozi book 46, “Geng zhu 耕柱” (SBCK 11/14b).
420.Mozi book 10, “Shang xian xia 尚賢下” (SBCK 2/18a).
198
virtue.
近者安其政，遠者歸其德。421
Rather than read Kongzi’s advice to the Duke of She as a stable Kongzi saying or Lunyu quotation, the frequency of the “please the near and attract the distant” cliché in the early literature
suggests that the Mozi attributed it to Kongzi precisely in order to accuse him of being a purveyor of hackneyed wisdom.
• The Hard Cases
Two additional corpora, the received Mengzi 孟子 and the Shanghai Museum bamboo manuscript collection, contain more than five Kongzi sayings with Lunyu parallels apiece (see appendices 2:T & 2:U):
Total Kongzi
attributions
Shanghai Museum manuscript corpus 78
Kongzi sayings with
Lunyu parallels
8 (10%)
Lunyu
attributions
0
Mengzi (Meng Ke 孟軻,
c. 372–289 BCE?)
12 (34%)
0
35
Although the percentage of Lunyu parallel sayings in these texts is still much lower than in late
Western Han, Xin, and Eastern Han texts, the Mengzi together with the Shanghai Museum (and
Guodian) “Zi yi 緇衣” (“Black Robes”),422 “Zhonggong 仲弓,” “Junzi wei li 君子為禮” (“The
noble man in his performance of ritual”), and “Dizi wen 弟子問” (“The disciples asked”) manuscripts are the best evidence we have for dating the Lunyu to the pre-Han period.423
421.Mozi book 35, “Fei ming shang 非命上” (SBCK 9/3b).
422.A “Zi yi” manuscript was also discovered among the Guodian materials. See Guodian Chu mu zhujian, pp.
127–137. However, in the present discussion I only consider the Shanghai Museum version.
423.Wojciech Simson (2006, pp. 36–39, and 2011) in particular has pointed to Mengzi 7B/37 as proof that the
Mengzi author was familiar with the Lunyu. Tsuda (1946, pp. 79–81) concluded that Kongzi sayings from the
199
But the evidence from the Shanghai Museum corpus is mixed, to say the least. On the one
hand, the language of six of its eight Lunyu parallel sayings (see Appendix 2:V, parallels 1, 2, 3,
6, 7) is not so far removed from the text of the received Lunyu. On the other, in no Shanghai Museum manuscript is a Lunyu parallel saying treated as particularly authoritative. In “Zi yi,” for instance, the Lunyu parallels are simply two among the twenty-three zi yue sayings in the text (see
Appendix 2:V, parallels 1 & 2). In “Dizi wen,” a Lunyu 14/35 parallel (“The Master said, ‘There
is no one who recognizes me!’” 子曰：莫我知也夫！ ) is followed by a question from Ziyou
instead of Zigong as it is in the Lunyu (Appendix 2:V, parallel 6).
The “Zhonggong” manuscript (Appendix 2:V, parallel 3) with its parallels to Lunyu 13/2 is a
more complicated case. Although the Shanghai Museum editors were only able to reconstruct a
portion of the manuscript, the surviving fragments reveal a text whose structure roughly parallels
that of the Shanghai Museum “Min zhi fumu 民之父母” (“Father and Mother to the People”)
manuscript discussed in chapter one (pp. 108–111), with Kongzi dispensing pithy advice that is
subsequently unpacked in the course of the dialogue. In “Zhonggong,” the initial instruction includes a fragment which parallels Lunyu 13/2: “treat the elderly as elderly and be kind to the
young. Put office-holders first, promote the worthy and talented, and forgive transgressions and
crimes” (老老慈幼，先有司，舉賢才，宥過與罪。). In both “Zhonggong” and Lunyu 13/2,
Zhonggong then asks Kongzi to clarify the nature of “promoting the worthy and talented” (ju
xian cai 舉賢才). But the manuscript version also seems to have elaborated the notions of “treating the elderly as elderly and being kind to the young” (lao lao ci you 老老慈幼), “putting office-holders first” (xian yousi 先有司), and “forgiving transgressions and crimes” (you guo yu
Mengzi were added to the Lunyu subsequent to its initial compilation.
200
zui 宥過與罪) without privileging any one of these elements over the other. The relative complexity, length, and detail of the “Zhonggong” version has led one scholar to characterize it as an
“unabridged” (fanben 繁本) version of the “abridged” (jieben 節本) Lunyu episode.424 I shall
have more to say about these parallels in chapter three (p. 217). For present purposes, it suffices
to point out that “Zhonggong” cannot have been a straightforward quotation of Lunyu 13/2.425
The evidence from the Mengzi is even less straightforward (see Appendix 2:W).426 Several
parallels seem to support the conclusion that the Mengzi author(s) was familiar with a Lunyu text.
Mengzi 2A/7 (parallel 2), 4A/14 (parallel 5), and 5B/7 (parallel 7), all of which are word-forword or nearly word-for-word Lunyu parallels, can be read as commentaries on Lunyu Kongzi
sayings, which might suggest that they were felt to be relatively authoritative. The most sustained commentary on a Lunyu Kongzi saying is Mengzi 7B/37 (parallel 8). The passage opens
with a question to Mengzi about a variant of Lunyu 5/22, and in his response Mengzi quotes variants of Lunyu 13/21 and 17/13 before quoting a variant of Lunyu 17/18 in order to explain the
meaning of the 17/13 variant.427 By invoking Lunyu Kongzi sayings to explain other Lunyu
Kongzi sayings, Mengzi 7B/37 provides the best evidence for the conclusion that the Mengzi was
compiled in a post-Lunyu milieu.428
424.Chen Tongsheng 2006.
425.Tsuda (1946, pp. 76–77) writes that his original impression was that the Mengzi postdated the Lunyu, but after
considering the evidence from Lunyu-Mengzi parallels he concluded that some Mengzi Kongzi sayings were sources
for the Lunyu.
426.See Zhao Zhenxin 1961 (pp. 16–20) for a parallel discussion of the evidence from the Mengzi. Pp. 17–19
includes a list of 32 Kongzi attributions in the text. Zhao also comes to the conclusion that the Lunyu had not yet
been compiled at the time of the Mengzi’s composition. Haupt (2006, pp. 53–70) finds 85 Kongzi mentions and 29
Kongzi attributions, of which only seven have parallels in the Lunyu. Haupt also counts a total of 13 passages in the
Mengzi with parallels to 16 Lunyu passages, with only three word-for-word parallels.
427.As an example of how subjective identifying parallels can be, Simson (2006, pp. 36–37) identifies more Lunyu
parallels in Mengzi 7B/37 than I have.
428.For this argument, see Simson 2011 & 2006.
201
Can these examples be reconciled with the theory that the Lunyu was compiled in the Western Han period at the earliest? It bears repeating that the Mengzi and Shanghai Museum manuscript parallels are exceptional. Of the more than 400 Kongzi attributions I have counted in ostensibly pre-Han texts, only 37 (9%) might be considered Lunyu parallels, and a number of these
exhibit variants significant enough to rule out the possibility that their authors quoted straightforwardly from the Lunyu. A number of these sayings are also prefaced by zi yue instead of Kongzi
yue, an observation which raises the possibility that they are quotations of someone other than
Kongzi.429 But let us suppose that, instead of Kongzi yue or zi yue, pre-Han authors had introduced these quotations with the phrase Lunyu yue 論語曰, thereby enabling us to reconstruct a
pre-Han Lunyu from among the corpus of pre-Han Kongzi sayings. Even then we would only be
able to reconstruct fragments of thirty or so Lunyu entries, or approximately 500 (3%) of the
16,000 characters of the text. The macro view of pre-Han Kongzi quotations cautions against
automatically reading the Mengzi and Shanghai Museum manuscript parallels as quotations of
the Lunyu.
Also recall my discussion in chapter one of Mengzi 5A/4 and its handling of a supposedly
spurious Kongzi saying (pp. 117–119). Unlike Wang Chong in the Eastern Han, Mengzi did not
cite the Lunyu or any other authoritative collection in order to deny the saying’s authenticity. Instead, he attacked the saying on two fronts: first, he asserted that “it was not the saying of a junzi” (此非君子之言); second, he cited a substitute Kongzi saying to override the first. Strikingly,
the second, Mengzi-approved Kongzi saying has no parallel in the received Lunyu. Mengzi 5A/
4’s handling of spurious Kongzi material also stands in sharp contrast to his handling of a misin-
429.These are the Lunyu parallels in the Guodian and Shanghai Museum “Zi yi” manuscripts.
202
terpreted Shi quotation. Mengzi and his interlocutor disagree over how to interpret the Shi; neither figure questions the authenticity of the line in question. However, Mengzi’s critique of his
interlocutor’s Kongzi quotation reveals a lack of agreement over what Kongzi actually said, the
problem of interpretation simply does not arise for Mengzi as it did for Liu Xiang or Wang
Chong. This is not a problem that would have arisen had the Lunyu or any other collection of
Kongzi sayings been regarded as authoritative when Mengzi 5A/4 was composed. The author of
Mengzi 5A/4 apparently had no recourse to a fixed text which would have settled the dispute
once and for all. In other words, the Mengzi Kongzi seems to have had a reputation but not a
canon. Here we might also note Mengzi 1A/7’s assertion that Kongzi’s disciples did not transmit
any of his teachings about the hegemons Duke Huan of Qi 齊桓公 and Duke Wen of Jin 晉文公
even though that claim is blatantly contradicted by the zi yue sayings at Lunyu 14/15, 14/16, and
14/17.430
Differences in the attributions of a number of Lunyu parallel sayings in the Mengzi also seem
to point to a pre-Lunyu milieu. In Mengzi 2A/2 (parallel 1), a Lunyu 11/3 variant is attributed to
Mengzi’s student Gongsun Chou; in Mengzi 2B/9, a Zengzi saying from Lunyu 12/21 appears as
a Mengzi saying; in Mengzi 3A/2 (parallel 3), a Kongzi saying from Lunyu 12/19 is attributed to
Mengzi and a Kongzi saying from Lunyu 2/5 is attributed to Zengzi; in Mengzi 4B/29 (parallel
6), a Lunyu 6/11 parallel appears in an unattributed passage which prefaces a Mengzi comment;
in Mengzi 5B/7 (parallel 7), Lunyu 10/20 appears as Mengzi’s student Wan Zhang’s description
of Kongzi, whereas in Mengzi 2B/2 the same line appears as a li yue 禮曰 (The rituals say) quotation; and in Mengzi 7B/37 (parallel 8), a Kongzi saying at Lunyu 13/21 is attributed to Mengzi
430.On this point see also Makeham 1996, p. 100, n. 74.
203
as a description of Kongzi. Such discrepancies in the attributions of Lunyu parallels was not a
feature of Eastern Han quotation practice.
Also significant is the fact that the Mengzi does not appear to discriminate between Lunyu
Kongzi sayings and non-Lunyu Kongzi sayings. Consider Mengzi 3B/9 and 4B/18:
<3B/9>Mengzi said, “...When the world declined and the Way waned, wicked persuasions and
violent deeds arose. There were subjects who killed their lords and sons who killed their
fathers. Kongzi was fearful and made the Chunqiu. The Chunqiu [records] the affairs of the
son of heaven. That is why Kongzi said, “Those who understand me do so only because of the
Chunqiu; those who blame me do so only because of the Chunqiu.”
孟子曰：...世衰道微，邪說暴行有作，臣弒其君者有之，子弒其父者有之。孔子懼，作
春秋。春秋、天子之事也；是故孔子曰：知我者其惟春秋乎！罪我者其惟春秋乎！431
<4B/18>Xuzi said, “Zhongni often praised water, saying, “Water! Water!” What did he
apprehend in water?” Mengzi said, “It gushes forth from a spring, ceaselessly day and night,
advancing after filling the hollows, expanding across the four seas. Things with roots are like
this, and this is what [Kongzi] apprehended in water. If something has no roots, then in the
rains of July and August it fills the gutters and ditches but dries up while you stand there
watching it. Thus the noble man is ashamed if his reputation exceeds his actual substance.”
徐子曰：仲尼亟稱於水，曰：水哉，水哉！何取於水也。孟子曰：原泉混混，不舍晝
夜，盈科而後進，放乎四海。有本者如是，是之取爾。苟為無本，七八月之間雨集，溝
澮皆盈；其涸也，可立而待也。故聲聞過情，君子恥之。432
It is hard to imagine a more definitive description of Kongzi and his importance than the nonLunyu saying at Mengzi 3B/9, which happens to be one of the very few statements on authorship
in any text from the pre-Han period.433 Mengzi 4B/8 indicates that the Mengzi author did not reserve commentary for Lunyu Kongzi sayings alone. Indeed, the appearance at Mengzi 7A/24 of
another episode featuring Kongzi “observing water” (guan shui 觀水) suggests that the Mengzi
attached particular importance to this non-Lunyu Kongzi tradition (see p. 239 below).
The one passage in any ostensibly pre-Han text which seems to provide unequivocal evi-
431.SBCK 6/13a.
432.SBCK 8/6a.
433.Kern 2011.
204
dence for the existence of a pre-Han Lunyu is Mengzi 7B/37. That passage, the second-to-last
episode in the entire Mengzi, which features a number of Kongzi sayings with Lunyu parallels
(albeit with significant variants) in succession, two of which it treats as objects of commentary.
But one cannot overstate how truly exceptional Mengzi 7B/37 is within the pre-Han corpus,
which contains no other passage in which Lunyu Kongzi sayings are juxtaposed with other Lunyu
Kongzi sayings in this way. The very uniqueness of 7B/37 coupled with its placement in the latter half of Mengzi book seven, far and away the most heterogeneous book in the entire Mengzi
collection, is reason to suspect that it is a later addition.434 An even bigger problem is that Mengzi
7B/37 appears in a text whose pre-Han origins cannot be taken for granted. Pre-Han and early
Western Han texts preserve very few Mengzi quotations with which to verify the existence of a
Mengzi text, and even some of these quotations have no parallel in the received Mengzi.435 The
mix of Lunyu Kongzi sayings and non-Lunyu sayings in the Mengzi would also fit a scenario in
which the text continued to evolve over the course of the early period until its boundaries were
fixed by Zhao Qi 趙岐 (d. 201 CE) in the late Eastern Han. As we have seen, Zhao Qi consciously exploited the numerous parallels between the Mengzi and Lunyu, and between Mengzi and
Kongzi, in order to enhance the authority of his own text. Such a compiler would have had a
434.According to Erik Maeder’s (1992, pp. 27–28) “looseleaf ring binder” theory whereby an existing text could be
“rearranged, shortened or expanded” as subsequent compilers saw fit, we might expect the final sections of received
texts to be be relatively disorganized and heterogeneous. However, Matthias Richter (2011) has argued persuasively
that early Chinese book formats do not explain the content of early texts.
435.For example, although the Xunzi “Xing e 性惡” (“Human nature is bad”) chapter attributes to Mengzi the view
that human nature is good, the Mengzi yue 孟子曰 attributions themselves do not match up with the received
Mengzi. For these quotations, see Xunzi 23 (SBCK 17/3a, 17/6b, & 17/8a). In the Shiji, the only attested passage is
Mengzi 1A/1 (Shiji 74.2343). However, Sima Qian also says that he chose “not to discuss the traditions [of Mengzi
and other masters’]” (不論其傳) because “many people nowadays have their writings” (世多有其書). Shiji 74.2349.
For Mengzi sayings that were not included in the received Mengzi, see Li Diaoyuan’s 李調元 (1734–1803) Yi
Mengzi 逸孟子 (Congshu jicheng chubian vol. 502) and Ma Guohan’s 馬國翰 (1794–1857) Yuhan shanfang jiyi
shu pp. 1780–1830.
205
strong motivation to pepper his text with Lunyu Kongzi sayings and thereby cement its association with the Lunyu Kongzi. Another possibility is that the Western Han Lunyu compilers looked
to the Mengzi as a source of Kongzi material. Indeed, the Mengzi’s interest in weeding out spurious Kongzi material might have made it an extremely useful resource for those looking to establish an authoritative collection of Kongzi sayings.
Lunyu parallels outside the corpus of Kongzi sayings
This chapter’s survey of Kongzi quotation practice has covered thousands of Kongzi sayings in
dozens of texts spanning more than half a millennium of early Chinese history. But there remains
one final kind of evidence to consider: parallels to Lunyu Kongzi sayings which are not Kongzi
sayings themselves. As we have seen, relatively few Lunyu Kongzi sayings circulated as Kongzi
sayings in the pre-Han and early Western Han periods. But many more Lunyu sayings seem to
have circulated as unattributed material or as sayings attributed to other figures (see Appendix
2:X). Whereas only 36 Lunyu Kongzi sayings have parallels in pre-Han and early Western Han
Kongzi yue material, by my count roughly 150 Lunyu entries have parallels whose attributions
differ from the Lunyu’s.436
Could it be that these parallels are actually implicit Lunyu quotations? Eastern Han authors
routinely borrowed or alluded to Lunyu Kongzi sayings without citing Kongzi or the Lunyu, so it
is at least possible that some of the parallels listed in Appendix 2:X were inspired by the Lunyu
or a Lunyu precursor text. Indeed, a number of these sayings are attributed to Kongzi disciples or
436.Here I have only counted parallels in texts through the end of the Western Han.
206
to Kongzi partisans like Mengzi, who might be expected to have borrowed liberally from the
Master.
Such a theory accounts less well for Lunyu parallels in the Guanzi (x14 entries),437 Mozi
(x5),438 Zhuangzi (x4),439 or other texts not written in obvious sympathy with Kongzi. More significantly, the majority of the 150 parallels identified in Appendix 2:X are not attributed to anyone at all, an observation which points to the generic or anonymous nature of this material. As
Richard Schultz observed in his landmark study of quotation practice in the Biblical, Near Eastern, and Western literary traditions,
The comparative material makes it abundantly clear that it is precisely those parallels that
display close correspondence which often are not quotations but are rather formulaic,
idiomatic or proverbial in origin and whose wording is so well established that it resists
modification...440
As parallels in Eastern Han texts attest, Lunyu Kongzi sayings were eventually quoted as Kongzi
sayings in the latter part of the early period due of the Lunyu’s pre-eminence.441 But the same
cannot be said of the pre-Han period. By attributing Lunyu parallel sayings to other sources or to
no sources at all, pre-Han and early Han authors revealed their ignorance of—or, at the very
least, their indifference to—Kongzi’s or the Lunyu’s association with those sayings.
Conclusion: The Lunyu as Western Han text
437.These are Lunyu 1/2, 1/3, 1/16, 5/12, 9/9, 11/16, 12/2, 12/11, 12/16, 12/17, 13/16, 15/21, 15/30, and 17/24.
438.Lunyu 1/6, 1/15, 9/9, 13/20, and 13/29.
439.Lunyu 2/14, 5/19, 12/4, and 19/1.
440.Schultz 1999, p. 224. See also Schultz 1999, p. 215: “...[M]ost lists of alleged quotations should be greatly
reduced, eliminating proverbs, clichés, idioms, refrains, figures of speech and topoi.” Li Ling (2004, p. 204, n.3)
makes a similar point about the early Chinese context.
441.An excellent example of this trend is Lunyu 7/1, which is discussed in chapter three (p. 249–254).
207
This survey of Kongzi quotations from the late Eastern Han through the pre-Han era has revealed
an unmistakeable pattern. When late Western Han, Xin, and Eastern Han authors quoted Kongzi
they tended to quote the Lunyu, particularly in contexts like edicts and memorials in which a
Kongzi quotation needed to be authoritative. Post-Lunyu authors also used, referenced, praised,
elaborated, and commented on Lunyu Kongzi sayings more often than on non-Lunyu sayings.
Sometimes they even rejected Kongzi sayings precisely because they did not appear in the Lunyu. In contrast, when early Western Han and pre-Han authors quoted Kongzi they tended not to
quote the Lunyu or to privilege Lunyu sayings over non-Lunyu sayings. Even when they did credit Kongzi with sayings that resembled Lunyu Kongzi sayings, those quotations tended to vary
significantly from their Lunyu counterparts.
How are we to account for the disparity between earlier and later Kongzi quotation practice?
One possible solution was suggested by Wang Chong: the Lunyu really was compiled in the preHan period by Kongzi’s disciples but was subsequently lost at some point prior to the Western
Han. The biggest problem with this theory is its reliance on the legend of the “ancient script-”
(guwen 古文) Lunyu in twenty-one sections supposedly recovered from the walls of Kongzi’s
ancestral home in Lu.442 According to the legend, a cache of ancient script-texts was discovered
after Liu Yu 劉餘, King Gong of Lu 魯恭王 (r. 154–128), set out to demolish Kongzi’s house in
order to clear space for his own palace. (According to the [pseudo-]Kong Anguo 孔安國 preface
to the Shangshu 尚書, the king then called off the demolition after entering Kongzi’s house and
“hearing tones of metal and stone and silk and bamboo” [聞金石絲竹之音], i.e., the sound of
ancient music.443) However, Tsuda Sōkichi and Kaneto Mamoru has argued persuasively that re-
442.See Appendices 2:D and 2:E for these references.
443.SBCK 1/3a. See also Hanshu 53.2414.
208
ports of an ancient script-Lunyu are likely the product of the late Western Han or Eastern Han
context.444 Not only does the roughly contemporaneous Shiji make no mention of an ancient
script-Lunyu, Liu Xiang’s son Liu Xin in a memorial dated to ca. 25 BCE also fails to mention
the Lunyu in his seemingly exhaustive list of “ancient script-” (guwen 古文) texts discovered in
the walls of Kongzi’s house.445 Kaneto also notes a number of discrepancies among the multiple
versions of the ancient-text Lunyu legend. For instance, while the “Yiwen zhi” dates the incident
to the “latter part of Emperor Wu’s reign” (Wudi mo 武帝末) (r. 141–87),446 the Hanshu biography of Liu Yu records that he was appointed King of Lu in 154 BCE during the reign of Emperor
Jing 景帝 (r. 156–141) and died twenty-eight years later in 128 BCE, a mere thirteen years into
Emperor Wu’s fifty-four year reign.447
Extant accounts of the discovery of an ancient script-Lunyu are too fanciful to be read as historical fact.448 It is tempting to speculate (as Kaneto does) that Ban Gu or someone like him expanded the legend of the “ancient script” texts to include the Lunyu, probably in order to enhance
the text’s authority by establishing its ancient provenance.449 Such a theory would also explain
444.Tsuda 1947, pp. 97–101, and Kaneto 1973.
445.Hanshu 36.1969. In this memorial Liu Xin urges Emperor Cheng 成帝 (33–7 BCE) to establish an official chair
for the study of the Zuozhuan 左傳. The only ancient script-texts it mentions are “remnant ritual texts” (Yi li 逸禮)
in thirty-nine sections and Documents (Shu 書) texts in sixteen sections.
446.This line can be found in the “Yiwen zhi” section on the Documents. See Hanshu 30.1706.
447.Hanshu 53.2414. In chapter 81 of the Lunheng, the section on the Lunyu dates the incident to Emperor Wu’s
reign but the section on the Shangshu dates it to Emperor Jing’s.
448.For a discussion of various problems with the legend, see Pelliot 1916, cited in Nylan 1994, p. 88 n. 14. See
also Nylan 1994 for the argument that the gu wen designation was far less meaningful in the Han period than it has
been for modern commentators, and for the conclusion that “[i]t is difficult to determine what gu wen and jin wen
(modern script) meant to Han intellectuals precisely because extant Han texts provide us with so little relevant
information” (p. 91).
449.Kaneto 1973. Nylan 1994, p. 88: “[T]he list of books associated with [Kong Anguo 孔安國 and King Xian of
Hejian 河間獻王, who supposedly played major roles in the recovery of ancient-script texts] grows ever longer over
time in a most suspicious fashion.” According to a fragment of Liu Xiang’s original Catalogue (Bie lu 別錄)
preserved in the preface to the Zhengyi 正義 (Correct Meaning) sub-commentary to the Shangshu, “At the end of
Emperor Wu’s reign one of the common people obtained a copy of the “Great Oath” [Documents text] from inside
209
why the “Yiwen zhi” lists several experts for the Qi and Lu Lunyu recensions but not a single expert for an ancient script-Lunyu.
Another possible explanation of the Kongzi quotation patterns observed in this chapter is that
the Lunyu (or parts of it) was compiled by Kongzi’s earliest disciples but was transmitted esoterically in the pre-Han period before being popularized during Emperor Wu’s reign.450 In this view,
the Lunyu really did exist in the early Warring States period but was not accessible to pre-Han
authors (with the possible exception of the Mengzi author[s]). Although nothing in this chapter
entirely precludes the esoteric transmission theory, I see little evidence that Kongzi’s teachings
were transmitted only within a small group of Kongzi disciples.451 As I argued in chapter one, extant Kongzi sayings from pre-Han and early Western Han texts reveal Kongzi quotation practice
to have been a widespread phenomenon, and the function of many of these sayings as dependent
comments also seems to point to exoteric applications. Texts like the “Kongzi shilun” and Lüshi
chunqiu treated Kongzi yue sayings as venues for the re-performance of received wisdom, and
particularly the wisdom found in widely circulating canonical traditions. Kongzi sayings do not
present themselves as closed teachings to be passed down verbatim from one generation to the
next. Moreover, the early literature preserves countless instances in which Kongzi and his disciples shared Kongzi sayings with rulers, officials, rivals, and various other outsiders. Presumably,
the walls [of his home] and submitted it to the throne. [Emperor Wu] gave it to the Erudites to transcribe and
explain, which took them several several months to complete. It was then circulated in order to instruct others” (武
帝末，民有得泰誓書於壁內者，獻之。與博士使讀說之，數月皆起，傳以教人。). This much more plausible
story might have been the inspiration for the King Gong legend.
450.Csikszentmihalyi 2001, p. 245: “During this gap [between Kongzi’s death and the biographical accounts of the
Han], narratives about the sage were passed down from master to disciple as part of an esoteric transmission
centered in Confucius’s temple in the small state of Lu.”
451.The Shanghai Museum “Wu wang jian zuo 武王踐阼” (“King Wu ascended the throne”) manuscript provides
one model of an esoteric text with its description of a “cinnabar writings” (dan shu 丹書) text and the various rituals
that must be performed before King Wu could be exposed to its teachings. See Shanghai bowuguan cang Zhanguo
Chu zhushu vol. 7, pp. 148ff.
210
esoteric knowledge would not have had such an exoteric presentation. As Sima Qian wrote in the
“Kongzi shijia,”
The patterned expressions Kongzi uttered while serving as an official or hearing cases could
be shared with others—there was nothing exceptional about them. As for the Chunqiu, when it
was appropriate to write he wrote and when it was appropriate to delete he deleted, so that
Zixia’s followers could not add a single word. When he instructed his disciples in the
Chunqiu, he said, “Later generations will know me by the Chunqiu, and they will also
condemn me by the Chunqiu.”
孔子在位聽訟，文辭有可與人共者，弗獨有也。至於為春秋，筆則筆，削則削，子夏之
徒不能贊一辭。弟子受春秋，孔子曰：後世知丘者以春秋，而罪丘者亦以春秋。452
Sima Qian’s implication is that only Kongzi’s judgments as encoded in the Chunqiu were a special brand of knowledge reserved exclusively for his disciples, the only people who possessed
the key to extracting Kongzi’s teachings from the text. Kongzi sayings, by contrast, were not
“exceptional” and “could be shared with others.” It is tempting to suppose that Sima Qian’s distinction between the Chunqiu and Kongzi’s “patterned expressions” (wenci 文辭) reflects an actual distinction between the truly esoteric Kongzi of the Chunqiu and the more accessible Kongzi
of Kongzi quotation practice.
A much simpler explanation for the disparity between earlier and later Kongzi quotation
practice is that the Lunyu simply did not exist prior to the Western Han period. Not only do we
find little to no evidence of the Lunyu’s existence in the pre-Han era, the remarkable diversity of
Kongzi quotations in the period reflects a milieu in which no source of Kongzi sayings was
deemed authoritative. Pinning down a precise date of compilation is a matter of speculation, especially given the very real possibility that the Lunyu evolved into its present form over the
course of the Western Han (see pp. 261–265). Nevertheless, the Lunyu’s apparent lack of influence on texts up to and including the Huainanzi, a text presented to Emperor Wu in 139 BCE,
452.Shiji 47.1944.
211
plus the fact that the first memorials (and possibly the first edicts) to quote the Lunyu date to the
reign of Emperor Wu (r. 141–87 BCE) at the earliest, plus the exclusive use of Lunyu Kongzi
sayings in the latest layer of the Shiji, suggest a date of compilation of ca. 150–130 BCE. In the
next chapter, I explore the implications of this conclusion for the study of the Lunyu itself.
212
CHAPTER THREE:
How the Lunyu was lun-ed: Nine Theories on the Compilation of the Analects
Accepting that the Lunyu did not exist prior to the early Western Han, the biggest obstacle to
reading it as an artifact of the Western Han is our ignorance of its compilers’ identity, their
circumstances, and their motivations. The Western Han was a time when more and more texts
were acquiring named authors and even postfaces which explained the circumstances of their
creation.453 Unfortunately, the Lunyu compilers were not so helpful. They do not tell us, for instance, whether they believed that their sayings accurately reflected the teachings of the historical Kongzi, or whether, like Zisi in the Kong congzi, they simply believed that their text “did not
fall short of [Kongzi’s] ideas” (不失其意) (see p. 107). Did they view their compilation as an act
of veneration towards an intellectual ancestor, or did they consciously “impute words” (yu yan
寓言) to Kongzi so that their audience would take their ideas more seriously? The text is silent
on these questions.
Aside from the content of the Lunyu itself, perhaps the only hint the Lunyu compilers left of
their intentions is the text’s title, “The Selected Sayings.” (This assumes, of course, that the title
was not added by some later editor. Wang Chong credits a certain Fu Qing of Lu 魯人扶卿, who
served as Inspector [cishi 刺史] of Jingzhou 荊州 during the latter part of Emperor Wu’s reign,
with having named the text.454). Parsing the yu 語 (“sayings” or “conversations”) of “Lunyu” is
unproblematic. But what did the Lunyu compilers mean to communicate when they characterized
453.See, e.g., the “Yao lüe 要略” chapter of the Huainanzi (book 20) and the “Zi xu 自序” of the Shiji (book 130).
454.Provincial inspectors (cishi 刺史) were first appointed in 106 BCE so Fu Qing likely reached this rank in the
latter part of Emperor Wu’s reign. On provincial inspectors, see Bielenstein 1980, p. 10. For my translation of Wang
Chong’s take on the Lunyu, see Appendix 2:F.
213
these sayings as having been lun 論-ed?455
In the previous chapter I presented a number of arguments against the Han bibliographers’
view of the Lunyu as a fifth-century BCE text. But the bibliographers nevertheless suggest an interpretation of lun which agrees with the big picture of Kongzi quotation practice. The “Yiwen
zhi 藝文志” records that Kongzi’s disciples “selectively compiled” (lun zuan 論篹) their respective records of Kongzi’s teachings, and Wang Chong writes that the text originally consisted of
“several dozen to a hundred sections” (數十百篇) which were whittled down to thirty by the early Western Han. In other words, both Ban Gu and Wang Chong conceived of the received Lunyu
as but a subset of an originally much larger corpus. Let us assume that a literate Western Han audience would have also grasped one of the main conclusions of chapter one, that Lunyu Kongzi
sayings represented only a small fraction of circulating Kongzi material in the period. In that
case, a Western Han reading of the Lunyu’s title might have been: “This is a text whose sayings
have been carefully selected (lun 論) from among the myriad sayings which have been, or might
possibly be, attributed to Kongzi and his disciples.”
The challenge for those who would decode the interests and motivations of the Lunyu compilers is to understand why these particular sayings were selected for inclusion in these particular
formats and in this particular arrangement.456 The only way to analyze Lunyu sayings as selections is to determine the range of options that confronted the Lunyu compilers and then compare
what was included with what was left out, an approach that requires careful attention to the evi-
455.Here I read lun 論 and yu 語 as a verb–object construction as opposed to noun-noun, i.e., as “discourses and
sayings” or perhaps “assessments and sayings.” However, this does not seem to be how the title was understood by
Ban Gu and Wang Chong, both of whom seem to have understood lun 論 as a verb.
456.On the notion of the Lunyu as a selection see Csikszentmihalyi 2001, Csikszentmihalyi 2002, pp. 144–45, p.
256, Csikszentmihalyi 2004, p. 30, and Li Ling 2004, pp. 298-299, 324.
214
dence from Lunyu parallels.457 If, as I have argued, the lun-ing of the Lunyu took place at some
point in the early Western Han, then our focus should be on parallels in texts which might have
been available to early Western Han compilers. Roughly contemporaneous texts like the Shiji,
Huainanzi, Hanshi waizhuan, and Liji will be especially invaluable sources insofar as a comparison with their authors’ handling of parallel Kongzi material might illuminate the editorial decisions which shaped the Lunyu.
Complications arise when we consider the ways in which a Kongzi saying might have been
lun-ed. As we have seen, the variability of Kongzi material in the period meant that a given
Kongzi saying could exist in multiple formats: as an independent saying, as part of a dialogue or
anecdote, or even as a third-person description of Kongzi. Once the Lunyu compilers determined
that a given saying was worthy of inclusion, they also had to decide whether to preserve the earlier context or to create a new one. The clustering of Lunyu sayings around certain themes and
tropes indicates that individual sayings were not viewed in isolation, and that the arrangement of
a Kongzi saying within the text was also a kind of lun-ing.458 Thus, internal evidence, and especially the arrangement and contextualization (or lack thereof) of Lunyu sayings, is a necessary
complement to the picture which emerges from textual parallels.
These considerations highlight the inadequacy of the phrase “the Lunyu compilers,” as if all
of the people ultimately responsible for the form and content of the received Lunyu were (a) en-
457.Contrast this approach with that of D.C. Lau (2000, p. 264): “It seems clear that even if we could restore the
Lunyu to the original form of any of the three versions [i.e., the Qi, Lu, and ancient text versions] we would be doing
no more than restoring it to what it was in the Western Han and that is unlikely to add very much to our knowledge
of the work. If our aim is to find out something about the composition of the Lunyu, the only possible approach is to
examine the text itself.”
458.Weingarten (2010, chapter two) in a study of Lunyu books one and seven has shown how close attention to the
arrangement of Lunyu Kongzi sayings might illuminate the sources that of individual Lunyu books. Hu Zhikui 1978
is a particularly useful structural analysis.
215
gaged in more or less the same activity at more or less the same moment in time, and (b) in
agreement with one another about the purpose(s) their text was supposed to serve. In this chapter, the term “Lunyu compilers” should be understood as referring to the unknown persons ultimately responsible for the form and content of the Lunyu as we have it today, either through
original composition, selective compilation, editing, arrangement, or rearrangement. I take it as
given that the heterogeneity of the Lunyu is a mark of multiple authorship, and I strongly suspect
that successive “Lunyu compilers” may have revised the text in ways that obscured the contributions of earlier “Lunyu compilers.”459
This exercise also has a number of limitations. The promise of Lunyu 4/15 and 15/3 to the
contrary, I do not believe that there is a single thread running through a text as disorganized and
unsystematic as the Lunyu. As we will see below, no single compilation strategy explains the
form and content of the Lunyu as we have it. A frustrating lack of evidence also precludes a unified theory of the Lunyu. While there are numerous Lunyu parallels in the early literature, enough
to hazard a few educated guesses about the compilers’ motivations, not all Lunyu sayings have
informative parallels. Most problematically of all, my effort to reconstruct the textual milieu of
the Lunyu compilers is necessarily tentative given our profound ignorance of the Western Han
context. Given these limitations, all conclusions in this chapter should be taken with a healthy
dose of skepticism. Less important for my purposes than proving any single argument is demonstrating what I believe is a more fruitful approach to the study of Lunyu Kongzi sayings, one that
privileges the evidence from textual parallels.
Finally, a caveat: the nine case studies of this chapter do nothing to advance the argument
459.On this point see also Loewe (ed.) 1993, p. 314: “The heterogeneous nature of the Lunyu as received precludes
the hypothesis that there was a single author.”
216
that the Lunyu is a Western Han text. To date I have not come across any set of parallels in
which the Lunyu version of a Kongzi saying considered in and of itself is obviously later than,
say, its Zuozhuan or Mengzi counterpart. My argument for reading the Lunyu as a Western Han
text rests solely on the evidence presented in chapter two, and my goal in the present chapter is
simply to demonstrate the plausibility of the revised dating by placing the text squarely within a
Western Han textual, intellectual, and political context. Thus, in the following pages I take it for
granted that the Lunyu was first put together in the Western Han in order to ask how we might
read the text profitably as a Western Han text. Below I outline nine possible answers to that
question, each of which highlights a different aspect of the lun-ing of the Lunyu.
• Section #1: Lunyu 13/2 and “promoting the worthy and talented”
In chapter two we encountered a close parallel between the brief dialogue at Lunyu 13/2 and the
much longer dialogue in the Shanghai Museum “Zhonggong 仲弓” manuscript (Appendix 2:V,
parallel 3). Both versions feature Kongzi dispensing advice to Zhonggong upon his appointment
as chief steward of the Ji 季 clan, one of three ministerial lineages which had come to dominate
the state of Lu 魯 in the late Spring and Autumn period. In the manuscript version, that advice
consists of four elements—“treating the elderly as elderly and being kind to the young, putting
office-holders first, promoting the accomplished and talented, and forgiving transgressions and
crimes” (老老慈幼，先有司，舉賢才，宥過與罪。)—which are then elaborated in the course
of the dialogue. But Lunyu 13/2 features only three elements in a different order—“putting office-holders first, forgiving lesser transgressions, and promoting the accomplished and talented”
(先有司，赦小過，舉賢才。)—of which only “promoting the accomplished and talented” is
217
expanded upon.
Judging from the close parallels in the wording and structure of the two texts, Lunyu 13/2
seems to have had a history which extended at least as far back as the late fourth century BCE
(assuming that estimates of the dating of the Shanghai Museum corpus are correct).460 Of course,
it is impossible to say for sure whether the Lunyu compilers had access to a text like the Shanghai Museum “Zhonggong” manuscript when they included Lunyu 13/2, or whether “Zhonggong”
had already been condensed into its “abridged” version (to borrow Chen Tongsheng’s term) prior
to the compilation of the Lunyu.461 But the comparison suggests that at some point someone,
quite possibly the Lunyu compilers, made a conscious decision to emphasize “promoting the accomplished and talented” by rearranging the core Kongzi saying and excising the explanations of
the other elements.462 The preservation of the other elements in Lunyu 13/2 as vestiges of the
longer, unabridged version is one hint that the Lunyu compilers in the Western Han were at least
aware of the earlier dialogue, and were perhaps even responsible for the abridgment. If so, then
in Lunyu 13/2 we have an example of the Lunyu compilers having expressed a clear preference in
the course of selecting from an earlier Kongzi tradition.
Why would the Lunyu compilers value “promoting the accomplished and talented” over the
other elements in the saying? Part of the answer to that question lies in the observation that Lunyu 13/2 is by no means the only entry in the text to indicate that “promoting” (ju 舉) or “recognizing men of accomplishment and talent” (zhi xian cai 知賢才) was of interest to the Lunyu
460.The dating of the Shanghai Museum manuscripts to ca. 300 BCE is supported by comparison with the Guodian
materials. For the dating of Guodian, see Li Xueqin 2000; for a dissenting view, see Wang Baoxuan 2000 and the
other papers cited in Allan 2009, p. 117, n. 6. See also Shanghai bowuguan cang Zhanguo Chu zhushu vol. 1, p. 2.
461.Chen Tongsheng 2006.
462.For similarly structured passages in the Lunyu see 12/21 and 20/2. In neither instance do enumerated elements
go unelaborated as they do in 13/2, an observation which highlights the truncated nature of Lunyu 13/2.
218
compilers. The term “understanding” or “recognizing others” (zhi ren 知人) appears in four sayings across four different books, in Lunyu 1/16, 4/7, 12/22, and 20/3. Lunyu 12/22 in particular
specifies that the most important kind of “understanding” (zhi 知) is “understanding others” (zhi
ren 知人). A number of other Lunyu sayings feature Kongzi commenting on the virtues, vices,
strengths, and weaknesses of his disciples and various historical figures. Such comments especially predominate Lunyu book 5, the first half of book 6, book 11, and the middle section of
book 14, and are prefaced by questions such as “what is So-and-so like?” (X也何如) (Lunyu 5/4,
5/8), “is So-and-so humane?” (ren 仁) (Lunyu 5/19, 6/30, 12/3, 14/1, 14/16, 14/17, 17/1), and
“can So-and-so be charged with the conduct of government?” (可使從政也與) (Lunyu 6/8). Still
other passages deal with the problem of identifying “teachers” (shi 師) (Lunyu 2/11), “stewards”
(zai 宰) (Lunyu 5/8), “noble men” (junzi 君子) (e.g., Lunyu 8/6 and 14/5), and those who “understand ritual” (zhi li 知禮) (Lunyu 3/15, 3/22, 7/31). Considered in isolation, such sayings are single-serve judgments on (from a Han perspective) long-dead individuals; taken together, they reveal a pervasive interest in the problems, the possibilities, and the rhetoric of character
evaluation generally.
Also significant is the way the Lunyu treats character as a decidedly transparent object of
contemplation, most obviously at Lunyu 2/10:
See how he acts, observe what he follows, and examine where he rests. How could the man
hide? How could the man hide?”
視其所以，觀其所由，察其所安。人焉廋哉？人焉廋哉？463
In chapter one I labeled a pair of Kongzi sayings from the Xici zhuan and the “Kongzi shilun”
manuscript “meta-commentaries” because they laid the theoretical groundwork for Yi and Shi in463.Stephen Owen (1992, pp. 19–20) opens his survey of Chinese literary thought with this saying, observing that
“Chinese literary thought began its development around this question of knowledge, a special kind of knowing as in
‘knowing a person’ or ‘knowing the conditions of the age.’”
219
terpretation, respectively (p. 82). There I characterized the sayings’ confidence in the readability
of the classics as a kind of commentarial conceit, without which the commentarial endeavor, and
the pursuit of traditional learning generally, is subject to the Zhuangzian criticism that such texts
are merely “the dregs of the ancients” (古人之糟魄).464 By asserting the essential readability of
character, Lunyu 2/10 likewise guarantees the value of learning how to “understand others” (zhi
ren 知人). Likewise, the cumulative effect of Kongzi’s character judgments is to present character evaluation as unproblematic, at least when conducted in the proper fashion. In this respect,
the Lunyu Kongzi differs from the Lüshi chunqiu Kongzi, who instructs his followers to remember that “knowing others is assuredly not easy” (知人固不易矣) after he mistakes Yan Hui for a
thief (p. 52).
It is probably no accident that Lunyu 2/10 parallels also appear in a range of sources identified by Matthias Richter as “characterological” texts, i.e., texts having to do with the evaluation
of official candidates.465 Another example is the phrase “crafty in speech and contrived in appearance” (qiao yan ling se 巧言令色) from Lunyu 1/3, 5/27, 17/17, and 15/25, a phrase which also
appears in the “Zengzi lishi 曾子立事” (“Zengzi establishes affairs”) and “Wen wang guan ren
文王官人” (“King Wen evaluates people for office”) chapters of the Da Dai Liji,466 the “Guan
ren 官人” (“Evaluating people for office”) chapter of the Yi Zhoushu 逸周書 (Remnant Zhou
Writings),467 and the “Lun ren 論人” (“Assessing others”) chapter of the Lüshi chunqiu,468 all of
464.Zhuangzi 13 (SBCK 5/34a).
465.Richter 2005. For Lunyu 2/10 parallels, see Yi Zhoushu 58, “Guan ren” (Yi Zhoushu jiaobu zhuyi 58.333), Lüshi
chunqiu 3/4, “Lun ren 論人” (SBCK 3/9a), Da Dai Liji 72, “Wen wang guan ren 文王官人” (SBCK 10/4a),
Huainanzi 13, “Fan lun 氾論” (SBCK 13/16b, in a passage prefaced by the line “thus the Way of assessing others...”
[故論人之道]), and Shuiyuan 16, “Tan cong 談叢” (SBCK 16/5a).
466.Da Dai Liji 49 (SBCK 4/4a) and Da Dai Liji 72 (SBCK 10/3b).
467.Yi Zhoushu jiaobu zhuyi 58.331.
220
which address the problem assessing official candidates. On the basis of these parallels, Richter
has argued persuasively that characterological traditions formed part of the Lunyu’s source
material.
Character evaluation in the Lunyu is as much a matter of being recognized as it is of recognizing others. Despite the Master’s injunction “not to worry that others do not know you” (不患
人之不己知) at Lunyu 1/16, 4/14, and 14/30,469 arguably the most persistent anxiety in the text is
that a talented and virtuous man will go “unrecognized” (bu zhi 不知) in his own time, e.g.,
when Kongzi laments at Lunyu 14/35 that “there is no one who recognizes me” (莫我知也夫).470
Lunyu 11/26 also depicts Kongzi teasing his disciples for constantly complaining, “I am not
known” (bu wu zhi 不吾知).471 It is tempting to imagine that the placement of Lunyu 1/1 at the
head of the text with its praise for the man who “does not resent going unrecognized by others”
(人不知而不慍) reflects the Lunyu compilers’ anxieties about recognition, the flip-side of their
interest in evaluating others.
Considered in and of itself, the Lunyu’s interest in character evaluation does not necessarily
point to a Western Han context. As we saw in chapter one, Kongzi comments on people constitute an entire sub-genre of Kongzi sayings which stretches back well into the pre-Han period.
However, resonances with an edict issued by Emperor Wen 文帝 (r. 180–157 BCE) in 165 BCE
hint that the representation of character evaluation in the Lunyu owes something to Western Han
political rhetoric surrounding the recommendation of talented individuals to the imperium. In the
468.Lüshi chunqiu 3/4 (SBCK 3/9a).
469.For a variant, see Lunyu 4/14.
470.On this theme, see also Lunyu 1/1, 1/16, 4/14, 9/28, 11/26, 13/2, 14/30, 14/39, 15/4, and 15/19.
471.See also Lunyu 14/39 and 15/4.
221
edict (Appendix 3:A), Emperor Wen ordered his officials to “select men of worth and integrity”
(xuan xian liang 選賢良) who were then expected to compose essays in which they would “fully
display their aims without hiding anything” (悉陳其志，毋有所隱). Emperor Wen also
promised to “personally survey [the responses] in order to discern how the grandees will assist
Us and whether or not they meet [Our criteria]” (朕親覽焉，觀大夫所以佐朕，至與不至。).
The language of the 165 BCE call for recommendations was echoed in similar edicts issued by
Emperor Wu in 134 and 130 and Emperor Yuan in 47.472
Compare the language of Emperor Wen’s edict with Lunyu 5/26 and 11/26 (Appendix 3:B),
both of which feature Kongzi enjoining disciples (Yan Yuan 顏淵 and Zilu 子路473 in 5/26 and
Zilu, Zeng Xi 曾皙, Ran You 冉有, and Gongxi Hua 公西華 in 11/26) to “each speak of [his]
aims” (各言其志).474 In Lunyu 11/26, the disciples’ speeches are then followed by Kongzi’s assessments of their qualifications for high office. Kongzi suggests that Zilu 子路 is not up to the
challenge of governing a small state, that Ran Qiu could indeed govern a small territory, and
Gongxi Hua’s seemingly modest desire to serve as a minor official of ritual matters actually
qualified him for higher office.
The appearance of similar anecdotes in two other Western Han compilations indicates that
such “interview scenes” (to borrow Oliver Weingarten’s term) were an important sub-genre of
Kongzi dialogue in the Han (see Appendix 3:C).475 These other versions also happen to exhibit
472.For Emperor Wu’s 134 BCE edict, see Hanshu 6.160–61; for the 130 edict, see Hanshu 58.2613–14; for
Emperor Yuan’s 47 BCE edict, see Hanshu 9.281–82 & 75.3171–72. All of these edicts feature the claim that the
emperor will “personally survey the responses” (qin lan yan 親覽焉) of recommended grandees.
473.Lunyu 5/26 in fact references a Jilu 季路, which seems to be an alternate name for Zilu.
474.This is the formulation in Lunyu 11/26. Lunyu 5/25 has a slightly different formulation: “Why don’t each of you
speak of your aims?” (盍各言爾志). For a similar although less well developed scene of evaluation, see Lunyu 5/8.
475.Weingarten 2010, pp. 156–160.
222
more definite parallels with Western Han edicts. Just as in Lunyu 11/26, disciples relate their
“aims” (zhi 志), after which Kongzi pronounces on their potential. In the Hanshi waizhuan book
9 parallel, Zilu is deemed “a brave man of service” (yong shi 勇士), Zigong “an eloquent man of
service” (bian shi 辯士), and Yan Yuan “a sagely man of service” (sheng shi 聖士), “a great man
of service” (da shi 大士), and a “fine” (mei 美) and “virtuous” (de 德) man. The overlap with the
Emperor Wen edict is most evident in the same episode from Hanshi waizhuan book nine in
which Kongzi says, “let each of you speak of your aims, and I shall survey them” (二三子各言
爾志，予將覽焉。). Here the use of the verb lan 覽 (survey), a term normally reserved for
kings, emperors, sages, or others in a position of political or intellectual authority, is striking.
While lan 覽 does not appear in the relevant Lunyu episodes, its use in the Hanshi waizhuan
speaks to the sub-genre’s contemporary appeal: Kongzi models the role of emperor, his disciples
would-be officials. The setting of the Shuiyuan and Hanshi waizhuan versions lends some support to this reading. According to the stele inscription commemorating the occasion, the Qin First
Emperor’s 秦始皇 while on an inspection tour of his newly unified empire ascended Mt. Tai 泰
山 in order to “survey [lan 覽] all around the eastern reaches [of the realm]” (周覽東極).476 Situating Kongzi interview scenes on mountaintops seems to have been a way to emphasize
Kongzi’s authority as a supreme judge of character, much as emperors sat atop the Han bureaucracy and personally judged (or so they claimed) the abilities of recommended officials.477
Taken together, the choice to highlight “promoting the accomplished and talented” in Lunyu
13/2, the large number of Lunyu sayings dealing with character evaluation, parallels with characterological texts, and resonances with Western Han edicts calling for the recommendation of tal-
476.Kern 2000, p. 19.
477.Whether any Western Han emperor personally oversaw the selection of capable officials is an open question.
223
ented officials to the imperium are compelling evidence for taking “recognizing men of worth
and talent” as one of the core concerns of the Lunyu compilers.478 The Lunyu’s perceived value
for the young Emperor Zhao and other scions of the Liu clan may have even been as a manual of
character evaluation, a skill they would have been expected to exercise once they had came of
age. On the flip side, aspiring officials could have read Lunyu Kongzi sayings as advice on becoming, or at least on presenting themselves as, “accomplished” and “talented” individuals for
the purpose of imperial recruitment.479 Attributing this characterological material to Kongzi,
whom Kongzi quotation practice had already established as a pre-eminent judge of character,
presumably lent them a significant amount of authority.
The case of Lunyu 13/2 is also an excellent illustration of the difficulties of dating Lunyu
Kongzi sayings. Lunyu 13/2 ultimately owes both its content and its structure to the person who
composed the earlier dialogue tradition represented by the “Zhonggong” manuscript. But the particular form of the Lunyu version with its focus on “promoting the accomplished and talented”
seems to reflect a concern with character evaluation in the context of Han political culture. The
basic ideas date to a pre-Han context, their selection and presentation—in a word, their lun-ing—
to a Western Han context.
• Section #2: Keeping up with the Laozi
478.The crucial importance of lun ren 論人 (assessing people) or lun chen 論臣 is a constant refrain in pre-Han and
early Western Han texts. See, e.g., the section beginning “the essentials of assessing others” (凡論人有要) in
Guanzi book 16, “Fa fa 法法” (SBCK 6/7a); the section beginning “the Way of assessing others” (論人之道) in
Huainanzi book 13, “Fan lun 氾論” (SBCK 13/16b); and the “Lun ren 論人” (“Assessing others”) section at Lüshi
chunqiu 14/3 (SBCK 3/8a).
479.For a parallel use of Kongzi to comment on the challenges of evaluating officials, see the “Ai gong wen wu yi
哀公問五義” (“Duke Ai asked about the Five Principles”) chapter of the Da Dai Liji, which begins with Duke Ai of
Lu asked Kongzi, “I wish to assess the men of service of my state and to hand over the running of my government to
one of them. How should I go about evaluating them?” (吾欲論吾國之士，與之為政，何如者取之？).
224
Another kind of selection strategy emerges from a comparison between Lunyu 2/17 and a set of
parallels from the Xunzi, Hanshi waizhuan, and Shuiyuan (see Appendix 3:D). Whereas Lunyu
2/17 is a standalone zi yue quotation, the three parallels incorporate that quotation into an elaborate scene of instruction featuring Kongzi and his disciple Zilu. All three non-Lunyu versions
open with Zilu paying a visit to Kongzi “in full dress” (sheng fu 盛服) and being criticized for
his ostentatiousness. Zilu exits and returns wearing simpler garb, after which Kongzi teaches him
the “essentials of speech” (yan zhi yao 言之要) and the “essentials of conduct” (xing zhi yao 行
之要), which are the keys to “humaneness” (ren 仁) and “knowledge” (zhi 知), respectively. In
the course of the lecture, Kongzi is made to utter a close parallel to the line from Lunyu 2/17.
The most compelling reason to read Lunyu 2/17 as a selection from an earlier dialogue is the
first six characters: “Zilu! I shall teach you about knowing” (由！誨女知之乎。). The vast majority of Lunyu Kongzi sayings directed at Kongzi disciples are prefaced by a disciple question
or some other piece of context that explains why that particular disciple is receiving the instruction. Strangely, Lunyu 2/17 makes no effort to explain why Zilu in particular is in need of this
lesson. But Kongzi’s mention of Zilu makes much more sense if we suppose that Lunyu 2/17 was
extracted from the earlier tradition represented by the Xunzi, in which case the first six characters
of Lunyu 2/17 might be read as vestiges of its pre-Lunyu history.480
One way to read Lunyu 2/17 as a selection is to ask: why did the Lunyu compilers choose
“taking knowing as knowing and not knowing as not knowing” (知之為知之，不知為不知)
over other lines in the original episode?481 Nothing in the Xunzi, Hanshi waizhuan, or Shuiyuan
480.Cheng Shude comes to the same conclusion at Lunyu jishi 2.110.
481.This is the Hanshi waizhuan version. The Xunzi substitutes yue 曰 for wei 為.
225
versions indicates that “taking knowing as knowing” was the most important element of
Kongzi’s instruction. To the contrary, “taking knowing as knowing” is balanced against “taking
being able as being able and being unable as being unable” (能之為能之，不能為不能), and
both lines set up Kongzi’s larger conclusions about “knowledge” (zhi 知) and “ability” (neng 能)
and then “wisdom” (zhi 知) and “benevolence” (ren 仁). For the Lunyu compilers, the saying at
Lunyu 2/17 seems to have trumped the anecdote’s core message—but why?
Unattributed parallels in the Xunzi and Hanshi waizhuan to the line “taking knowing as
knowing and not knowing as not knowing” suggest one answer.482 If this particular saying circulated as an independent proverb, then the Lunyu compilers might have simply extracted the element that most resembled something they had heard before. A more intriguing possibility is suggested by a set of parallels from the Laozi, Lüshi chunqiu, Zhuangzi, Huainanzi, and Wenzi (see
Appendix 3:E). Proverbs built around the idea that knowing about knowing is, to quote the
Laozi, the “highest” (shang 上) form of knowledge appear throughout pre-Han and early Han
texts, including in the Guodian and Mawangdui manuscripts.483 Of these parallels, the version in
§71 of the received Laozi (“Knowing that one does not know is highest; not knowing whether
one knows is a fault”) seems to have had a strong association with Laozi in the Western Han, at
least judging from a Laozi yue 老子曰 attribution in Huainanzi book twelve.484 Given the prevalence of Laozi manuscript finds and Laozi quotations in pre-Han and Western Han sources, my
suspicion is that the Lunyu compilers were familiar with Laozi §71 as a Laozi saying, in which
case Lunyu 2/17 might have appealed to them as Kongzi’s version of a well known proverb tradi-
482.See Xunzi book 8, “Ru xiao 儒效” (SBCK 4/17b & 4/18b) and Hanshi waizhuan book 5 (SBCK 5/4b)。
483.Mawangdui Han mu boshu, “De jing 德經,” p. 6.
484.SBCK 12/14b.
226
tion.485 Tellingly, the three characters appended to Lunyu 2/17 but absent in the parallels—“this
is knowing” (shi zhi ye 是知也)—indicate that illuminating a truer form of zhi 知 (“knowing” or
“understanding”) was indeed their primary concern.
*
*
*
Lunyu 2/17 is not the only Lunyu Kongzi saying that reads as a response to an earlier proverb tradition. Lunyu 14/34 quotes an anonymous proverb which it then goes on to criticize:
Someone said, “What of ‘repaying wrong with virtue’?” The Master said, “Then how shall we
repay virtue? Repay wrong with rectitude but repay virtue with virtue.”
或曰：以德報怨，何如？子曰：何以報德？以直報怨，以德報德。
As it happens, the saying coyly introduced by the phrase “someone said” (huo yue)486 also appears (with a minor syntactic variant) in received and manuscript versions of the Laozi:
Act without acting, work without working, taste without tasting. Make the small great and the
few many. Repay wrong with virtue.
為無為，事無事，味無味。大小多少，報怨以德。487
Several parallels in Han texts confirm that the phrase “repay wrong with virtue” circulated widely in the period (see Appendix 3:F).
In chapter one we saw that Kongzi comments typically expressed approval or disapproval for
another figure or text but sometimes also reworked or amplified a piece of text in order to draw
out its wisdom. Here, too, the zi yue saying does much more than simply evaluate the original
saying. By replacing the word de 德 (*tʕəәk) or “virtue” with the nearly homophonous zhi 直
(*N-t<r>əәk) or “rectitude,” Lunyu 14/34 piggybacks off of the earlier Laozi saying to create a
485.Robert Eno (2003, pp. 239–40, n. 2) assumes that Lunyu 2/17 was a Kongzi saying from its inception. Haupt
(2006, pp. 134–135) disagrees and suggests that Kongzi was simply the “spokesperson” (Sprecher) for what was a
common idea.
486.Lunyu 2/21, 3/11, 3/15, 3/22, 5/5, 5/11, 9/14, 14/9, 14/34, 14/44. Questions from anonymous interlocutors—the
“someone” of “someone said” (huo yue 或曰)—are relatively infrequent in the Lunyu. In every instance, they pose
questions whose premises are rejected by Kongzi.
487.Laozi 63 (SBCK 下/13a).
227
new saying, one that conceptually and aurally co-opts the original.
Sima Qian in the conclusion to his biography of Laozi had this to say about the relationship
between pro-Laozi and pro-Ru partisans in the Western Han:
Nowadays those who study Laozi criticize Ru learning and those who engage in Ru learning
also criticize Laozi. Isn’t this a case of “different Ways not conferring”? [Lunyu 15/40].
世之學老子者則絀儒學，儒學亦絀老子。道不同不相為謀，豈謂是邪？488
Against this background, Lunyu 14/34 might be read as an artifact of Western Han intellectual
factionalism, as the Lunyu compilers’ implicit criticism of the only other pre-Han master whose
authority and quotability rivaled Kongzi’s. Like the author of the “man of Chu lost his bow”
episode from the Lüshi chunqiu (p. 57), the Lunyu compilers seem to acknowledge a competition
between these two authorities and went out of their way to establish the supremacy of the one
over the other. In the Lunyu as in the Lüshi chunqiu, that competition plays out on the level of
rhetoric, as dueling sayings.
*
*
*
The desire to co-opt a pre-existing proverb tradition might have also motivated the selection of
Lunyu 9/26:
The Master said, “The three armies can be robbed of their leader, but a common man cannot
be robbed of his will.”
子曰：三軍可奪帥也，匹夫不可奪志也。
Parallels to Lunyu 9/26 in the Sunzi bingfa 孫子兵法 (Master Sun’s Art of War) and Zuozhuan
point to an early tradition of “military mottos” (jun zhi 軍志) built around the notion of “stealing
hearts” (duo xin 奪心) (Appendix 3:G). The saying’s appearance in the Sunzi bingfa is especially
noteworthy, the notion that an opponent is open to manipulation being a central conceit of a military manual in the same way that the readability of traditional texts is crucial to the rhetoric of
488.Shiji 63.2143.
228
traditional commentary. Why bother studying such techniques if one’s opponents are immune to
manipulation?
Considered against this backdrop, the wording of the first part of Lunyu 9/26—“The three
armies can be robbed of their leader”—might be read as a partial acknowledgement of the earlier
proverb tradition. Crucially, however, the Lunyu does not allow that tradition to speak for itself.
Instead, it limits its relevance only to the case of military leaders; the “will” (zhi 志) of the “common man” (pifu 匹夫), by contrast, cannot be taken away from him.489 Read as a comment on
military proverbs, then, the message of Lunyu 9/26 would seem to be: “Yes, there is a military
sphere within which scheming military men with their military manuals can be manipulated
against their will. But military men know nothing of true will and steadfastness.” In this respect,
Lunyu 9/26 can be read in conjunction with the episode at Lunyu 15/1 in which Kongzi refuses to
answer a question about “[military] formations” (chen 陳) by insisting on the primacy of ritual.
Once again, the Lunyu compilers seem to manifest their biases on the level of rhetoric, undercutting proverb traditions associated with groups to which they were hostile.
*
*
*
The final example of a Lunyu saying chosen for selection because of pressure from competing
wisdom traditions is Lunyu 14/35, which features Zigong asking Kongzi to clarify a lament:
The Master said, “There is no one who understands me!” Zigong said, “How is it that no one
understands you?” The Master said, “I do not resent heaven and I do not blame men. At worst
I learn [whatever I do not know] and at best I achieve comprehension. If anyone understands
me it is surely heaven!”
子曰：莫我知也夫！子貢曰：何為其莫知子也？子曰：不怨天，不尤人，下學而上達。
知我者其天乎！
489.Another set of parallels in the Liji (Appendix 3:I), especially those in “Zi yi 緇衣” (“Black robes”), might
suggest that the second part of Lunyu 9/26 was inspired by a separate proverb tradition concerning the inviolability
of a person’s “will” (zhi 志). If so, then Lunyu 9/26 might be read as a synthesis of two competing proverb
traditions.
229
Here as well the core saying (“There is no one who knows me”) has parallels in a number of early texts (Appendix 3:H). While variants of the “no one knows me” trope appear in the Shi corpus, the precise phrase “there is no one who knows me” is also found in the received and
Mawangdui versions of the Laozi as well as in the (very fragmentary) “Dizi wen 弟子問” (“The
disciples asked”) manuscript from the Shanghai Museum corpus. However, Kongzi’s exclamation in “Dizi wen” is followed by a question from Ziyou 子游 and not Zigong as it is in the
Lunyu.
The “no one understands me” lament seems to have acquired a special significance in the
Western Han thanks to its appearance in the “Li sao 離騷” (“Encountering sorrow”) and related
texts. Traditionally ascribed to Qu Yuan 屈原, the “Li sao” is the first and most influential poem
in the Chuci 楚辭 (Verses of Chu) collection. Its coda (luan 亂) includes the line “there is no one
in my state who knows me” (國無人莫我知兮). Parallel instances in “Huai sha 懷沙” (“Clutching sand”), another Chuci piece also attributed to Qu Yuan, as well as in Jia Yi’s “Diao Qu Yuan
fu 弔屈原賦” (“Mourning for Qu Yuan”), testify to the idea’s importance within a prominent
genre of Han literature (see Appendix 3:H). Given this context, Lunyu 14/35 might be read as an
effort to establish Kongzi’s ownership of a trendy literary trope, one that (once again) had a long
history in the Laozi and elsewere.490
*
*
*
The first section of this chapter framed the Lunyu compilers as creatures of a Western Han political milieu. Sensitive to their audience’s interest in the imperial recruitment process, they peppered their text with Kongzi sayings having to do with the evaluation of character and talent. In
490.For other Chuci resonances, see p. 240 below. Given these parallels, it is entirely possible that the mysterious
Old Peng 老彭 of Lunyu 7/1 is the same Peng Xian 彭咸 invoked in the coda to the “Li sao 離騷.”
230
contrast, the analysis of this section has cast the compilers as collectors of proverbs. As proverb
collectors, they must have felt some pressure to include sayings that were both memorable and of
interest to their audience. As heirs to a tradition of proverbial language shaped in part by widely
circulating texts like the Laozi, the Lunyu compilers modeled many of their sayings on earlier
proverb traditions. The analysis of this section points to a desire to match those traditions at the
very least, and perhaps even to supplant them in an effort to present Kongzi as a font of supreme
yet eminently recognizable wisdom.
• Section #3: Lunyu 2/19 & 14/21: white-washed sayings
Yet another lun-ing strategy emerges from a comparison between Lunyu 14/21 and a close parallel in the Zuozhuan entry for the fifteenth year of Duke Ai’s reign (r. 494–477 BCE) (Appendix
3:I). Both texts relate Kongzi’s response to Chen Chengzi’s 陳成子 assassination of Duke Jian
of Qi 齊簡公 in 480 BCE while Kongzi was serving Duke Ai of Lu 魯哀公; both feature Kongzi
advising the Duke to launch a punitive attack on Qi; both have Duke Ai instructing Kongzi to
seek permission for the attack from the powerful ministers of Lu; and both conclude with nearly
identical Kongzi sayings (“I am ranked after the grandees so I did not dare not to inform you” 以
吾從大夫之後，不敢不告也).491
But there is a crucial difference between the two versions. The Zuozhuan (like the much later
Kongzi jiayu)492 relates that Kongzi simply “refused” (ci 辭) Duke Ai’s order to report to the
head of the Jisun clan and promptly resigned (tui 退) his post. In contrast, the Lunyu relates that
491.The Zuozhuan has yan 言 (speak) where the Lunyu prefers gao 告 (report).
492.Kongzi jiayu 41, “Zheng lun jie 政論解” (SBCK 9/26a–b).
231
Kongzi reluctantly obeyed the order: “[Kongzi] went to the three ministers and reported, but they
did not allow [the attack]” (之三子，告，不可。). Whereas the concluding saying in the
Zuozhuan (“I am ranked behind the grandees and so I did not dare not to speak) justifies
Kongzi’s resignation and is addressed to unnamed “others” (ren 人), in the Lunyu it is addressed
to Duke Ai himself and expresses Kongzi’s displeasure with his lord. The Lunyu Kongzi is then
made to utter the saying again, apparently to the “three ministers” (san zi 三子) after they also
refuse his advice.
What explains this discrepancy? Assuming that the Zuozhuan is indeed the earlier text, my
suspicion is that the Lunyu compilers edited the episode in such a way as to reject the portrayal
of Kongzi as an official who willingly disobeyed his lord’s direct command.493 Elsewhere in the
Lunyu one encounters sayings to the effect that not “disobeying” (wu wei 無違) is the essence of
filial piety (Lunyu 2/5),494 that the truly filial are “loyal” (zhong 忠) (Lunyu 2/20),495 and that filial men are not “fond of going against superiors” (hao han shang 好犯上) (Lunyu 1/2). If the
Lunyu compilers were interested in portraying, it seems they could not have included the
Zuozhuan episode in its original form. Their solution was to excise Kongzi’s resignation and replace it with the line “[Kongzi] went to the three ministers and reported, but they did not allow
[the attack]” (之三子，告，不可。), thereby representing Kongzi as a disgruntled but still obedient employee. They also preserved the saying “I am ranked after the grandees and so I did not
dare not to inform you” as an expression of Kongzi’s displeasure with Duke Ai, and then (per-
493.Cheng Shude (Lunyu jishi p. 1000) quotes Fang Guanxu’s 方觀旭 Lunyu ouji 論語偶記 to make the point that
Kongzi’s behavior in the Lunyu but not the Zuozhuan was zheng 正 (correct) precisely because “the lord’s command
must be accepted” (jun ming bu ke bu feng 君命不可不奉).
494.See also Lunyu 4/18 on not “disobeying” (wei 違) one’s parents.
495.On the definition of zhong 忠 in the Lunyu and in early texts generally, see Goldin 2008.
232
haps somewhat clumsily) repeated the saying as an expression of Kongzi’s displeasure with the
three ministers. In the Zuozhuan version, the final saying signals that Kongzi prized principle
above loyalty; in Lunyu 14/21, it underscores Kongzi’s unwavering sense of duty in service of a
weak-willed lord.
*
*
*
The desire to whitewash a morally problematic Kongzi tradition might have also motivated the
selection of Lunyu 2/19, whose Kongzi saying has a parallel in Huainanzi book sixteen (Appendix 3:J). The Huainanzi and Lunyu versions agree on two points: the core saying’s contrast between “raising the straight against the crooked” and “raising the crooked against the straight” and
its association with Kongzi’s official career in Lu.496 But whereas the Huainanzi casts the saying
as Kongzi’s justification for “conforming to [the usurping Jisuns’] conduct before entering their
government” (順其所為，而後與之入政), the Lunyu version is addressed to Duke Ai and
makes no mention of the Jisuns. Significantly, its advice to “raise the straight and set it against
the crooked” acknowledges none of the moral complexity of the Huainanzi version. The one
Kongzi readily acknowledges the need to engage in “crooked” endeavors for the greater good,
the other simplistically asserts the need to always confront the bad with the good.
Aside from the theme of “raising the straight”/“raising the crooked,” the shared use of *-əәk
and *-aŋʔ rhymes would seem to imply some sort of genetic relationship between them. Whether
the one was an adapted quotation of the other or whether both versions were quoting some unknown source, my suspicion is that the more regular rhyme scheme of the Huainanzi version is
the earlier one, and that the Lunyu compilers sacrificed that scheme in the course of transforming
496.A variation on the same phrase also appears at Lunyu 12/22 and in the “Kongzi shijia” (Shiji 471935: “Raise the
straight against the crooked and the crooked will be straight” [季康子問政，曰：舉直錯諸枉，則枉者直。])
233
it from a statement about ethical gray areas, about laudable ends justifying less-than-moral
means, into a black-and-white statement against compromising one’s morals.
Although it does not exhibit verbal parallels to the Lunyu and Huainanzi passages, a Kongzi
anecdote from the Lüshi chunqiu speaks to an interest in assessing the propriety of Kongzi’s service to the Ji clan:
The head of the Ji clan had usurped the power of the ducal house of Lu. Kongzi wanted to
instruct him in the proper methods but [knew that] he would be perceived as an outsider, so he
accepted a salary and took the opportunity to offer his persuasions. When people in Lu
criticized him for this, Kongzi responded, "The long-dragon eats and swims in clear water; the
chi-dragon eats in clear water but swims in muddy water; fish eat and swim in muddy water.
Now, I have not ascended to the level of a long-dragon but I have not descended to that of
fish. I am perhaps a chi-dragon!” When you want to accomplish something, how do you
coincide perfectly with the carpenter's line? Rescuing the drowning requires getting wet;
chasing someone running away means having to sprint.
季孫氏劫公家。孔子欲諭術則見外，於是受養而便說。魯國以訾，孔子曰：龍食乎清而
游乎清，螭食乎清而游乎濁，魚食乎濁而游乎濁。今丘上不及龍，下不若魚，丘其螭
邪。夫欲立功者，豈得中繩哉？救溺者濡，追逃者趨。497
Although expressed in different terms, the Lüshi chunqiu Kongzi’s embrace of morally problematic means—the “muddy water” (zhuo 濁) of serving the Ji clan—for righteous ends—the “clear
water” (qing 清) of advising as to the “proper methods” (shu 術)—is much closer in spirit to the
Huainanzi Kongzi’s apology. By adapting the language of the Huainanzi tradition and also jettisoning its context, the Lunyu compilers may have sought to remove any hint of impropriety on
Kongzi’s part. The end result was a purer, if less interesting, portrait of Kongzi.
Why would the Lunyu compilers have white-washed Kongzi traditions in this way? Here we
might recall that the earliest accounts of anyone actually using a Lunyu text (Sima Qian’s comment on the “Zhongni dizi liezhuan” excepted) describe it as a textbook for Han royal princes.
Assuming that the text was compiled or edited with that use in mind, the Lunyu’s rejection of
497.Lüshi chunqiu 19/8, “Ju nan 舉難” (“The difficulty of promoting”). Translation adapted from Knoblock &
Riegel 2000, pp. 505–506.
234
ethical gray areas might reflect the pedagogical motivations of the Lunyu compilers. A Kongzi
who embraced the “crooked” or “muddy water,” even if only as a means to an end, was perhaps
felt to be an inappropriate model for impressionable future emperors and kings.
• Section #4: Epitomizing Kongzi sayings
One of the most widely attested Kongzi anecdotes in pre-Han and Western Han literature relates
Kongzi’s conversations with his disciples while stuck “between Chen and Cai” (Chen Cai zhi
jian 陳蔡之間).498 One index of the story’s popularity in the period is that it was also appropriated by critics who sought to subvert the story to portray Kongzi as an unrealistic dreamer who
brought his troubles on himself (Zhuangzi) or as a rank hypocrite (Mozi). Still other authors simply referenced the story as if it was common knowledge. For instance, Liu Sheng 劉勝, a son of
Emperor Jing 景帝 (r. 156–141 BCE) who reigned as King of Zhongshan 中山王 for 42 years (r.
154–112 BCE), alluded to the story in a memorial dated to 138 BCE: “King Wen was detained in
Youli and Kongzi was trapped between Chen and Cai” (文王拘於牖里，孔子厄於陳蔡。).499
In chapter one (pp. 29–31) I noted that extant versions of the legend generally share a similar
framework but differ with respect to the Kongzi sayings themselves. So long as an author stuck
to a few basic points, e.g., that Kongzi and his disciples were “surrounded” (wei 圍) by hostile
forces and that they had little or no food, he was apparently free to construct a conversation
around those details in whatever way he saw fit. How, then, did the Lunyu compilers deal with
the “between Chen and Cai” anecdote tradition? Which, if any, of the story’s elements did they
498.For my earlier discussion of Chen and Cai narratives in the context of Kongzi quotation practice, see pp. 29–31.
499.Hanshu 53.2423. For a similar example, see Mengzi 7B/18 (SBCK 14/6b).
235
accept? And what do those choices reveal about the compilers’ motivations?
The Lunyu happens to include a number of sayings which reference Kongzi’s misadventures
between Chen and Cai. Three of these sayings mention Chen and/or Cai explicitly:
Lunyu 5/22 While in Chen the Master said, “Let us go home! Let us go home! Our young men
are wildly ambitious and have great accomplishments for all to see, but they do
not know how to prune themselves.”500
子在陳，曰：歸與！歸與！吾黨之小子狂簡，斐然成章，不知所以裁之。
Lunyu 11/2 The Master said, “None of those who followed me in Chen and Cai ever reached
my gates.”
子曰：從我於陳、蔡者，皆不及門也。
Lunyu 15/2 While in Chen [the Master] ran out of food and his followers were exhausted and
unable to rise. Zilu had a resentful look and said, “Does even a noble man hit rock
bottom?” The Master said, “The noble man is at rock bottom as a matter of course.
When a petty man hits rock bottom he loses his scruples.”
在陳絕糧，從者病，莫能興。子路慍見曰：君子亦有窮乎？子曰：君子固
窮，小人窮斯濫矣。
Of the three passages, Lunyu 15/2 speaks most directly to the Chen and Cai legend insofar as it
describes the starvation and exhaustion suffered by Kongzi and his disciples. While Lunyu 5/22
and 11/2 do not include any narrative details,501 they do happen to echo Kongzi’s criticisms of
certain disciples in other Chen and Cai anecdotes, e.g., when the Kongzi of the Lüshi chunqiu
version calls Zilu and Zigong “petty men” (xiao ren 小人) for questioning Kongzi’s good cheer
in such dire circumstances.502
To the above list we can add three more sayings whose parallels suggest an association with
the Chen and Cai legend:
500.Translation after Lau 2000, p. 44.
501.Cheng Shude (Lunyu jishi pp. 739–41) reads Lunyu 11/2 against Mengzi 7B/18 (“When the junzi was in dire
straits between Chen and Cai, he had no associates above [i.e., among the political elites] or below [i.e., among the
common people]” (君子之厄於陳蔡之間，無上下之交也。).
502.Lüshi chunqiu 14/6 (SBCK 14/17b)
236
Lunyu 9/28 The Master said, “Only after the year grows cold does one know that the pine and
cypress are the last to wither.”
子曰：歲寒，然後知松柏之後彫也。
Lunyu 15/3 The Master said, “Ci, do you suppose that I know what I do because I have studied many kinds of things?” He replied, “I do. Is that not the case?” “It is not. I
bind it all together with a single [thread].”
子曰：賜也，女以予為多學而識之者與？對曰：然，非與？曰：非也，予一
以貫之。
Lunyu 15/6 Zizhang asked about following [the Way]. The Master said, “If your words are
loyal and trustworthy and your actions sincere and respectful, then you will follow
it even in foreign lands. But if your words are not loyal and trustworthy and your
actions are not sincere and respectful, then will you follow it even in your local
area? While standing you must see it facing you, while riding you must see it on
the handle-bar. Only then will you follow [the Way].” Zizhang wrote this down on
his sash.
子張問行。子曰：言忠信，行篤敬，雖蠻貊之邦，行矣。言不忠信，行不篤
敬，雖州里，行乎哉？立則見其參於前也，在輿則見其倚於衡也，夫然後
行。子張書諸紳。
Lunyu 15/3 follows Lunyu 15/2 in the “Kongzi shijia” biography, the implication being that
Zigong and Kongzi had this exchange between Chen and Cai.503 Likewise, the “Zhongni dizi
liezhuan” prefaces Lunyu 15/6 with the line, “Another day, when [Zizhang] followed Kongzi between Chen and Cai and encountered difficulties, he asked about following [the Way]” (他日從
在陳蔡閒，困，問行。).504 However, the Shiji is the only early text I am aware of which contextualizes Lunyu 15/3 and 15/6 as statements uttered between Chen and Cai, and the Kongzi
sayings themselves do not seem to depend on that context.
Variants of Lunyu 9/28, on the other hand, appear as Kongzi sayings within much longer
Chen and Cai anecdotes from the Lüshi chunqiu and Zhuangzi (as well as in unattributed pas-
503.Shiji 47.1930.
504.Shiji 67.2204.
237
sages in the Xunzi and Huainanzi) (Appendix 3:K). Thus it would seem that in Lunyu 9/28 we
have another example of a Kongzi saying extracted from an earlier anecdote tradition. Above I
noted that the Kongzi saying at Lunyu 2/17 was a curious selection because the notion of “knowing that one knows” was neither the most important nor the most representative element of the
Kongzi dialogue found in the Xunzi, Hanshi waizhuan, and Shuiyuan. This led me to surmise that
the Lunyu compilers cared little for the longer dialogue and simply extracted the discrete idea
they were most interested in. However, Lunyu 9/28 exhibits an altogether different relationship
to its source material, the memorable metaphor of the pine and cypress succinctly expressing the
idea that true virtue manifests itself only in the direst of circumstances. Unlike Lunyu 2/17, the
selection of Lunyu 9/28 reflects a sensitivity towards its source material, and perhaps even an interest in capturing its core lesson in a manner similar to Kongzi comments on anecdotes in other
texts. A compiler who set out to extract a representative standalone Kongzi saying from these
Chen and Cai narratives could do far worse than Lunyu 9/28.
Seen from this perspective, Lunyu 15/2 might also be read as an expression of another core
message of the Chen and Cai legend: since “the noble man is at rock bottom (qiong 窮) as a matter of course,” there is no need for Kongzi or his disciples to despair of their situation. The fact
that the word qiong 窮 (to be at rock bottom) appears in several extant versions indicates that it
was one of the key elements of the legend, in which case the Lunyu compilers might have isolated the term and then built a new saying around it.505 In the process, they epitomized a popular legend with a pithy saying much as the Kongzi comments discussed in chapter one epitomized various kinds of texts and (quasi-)historical figures. Put simply, Lunyu 15/2 might be said to express
505.In contrast, John Makeham (1998, p. 76) concludes from the shorter length of Lunyu 15/2 that it predates the
more elaborate versions in the Mengzi, Zhuangzi, and elsewhere.
238
the “greater meaning” (da yi 大義) of the Chen and Cai legend.
*
*
*
Another Lunyu Kongzi saying with a number of precedents in an earlier anecdotal tradition is
Lunyu 9/17:
The Master was by a river when he said, “Such is its current! It stops neither day nor night.”
子在川上曰：逝者如斯夫！不舍晝夜。
Stories about Kongzi’s fascination with rivers appear in several early texts, including twice in the
Mengzi and once each in the Xunzi, Shizi, Zhuangzi, Shiji, Da Dai Liji, Shuiyuan, Kongzi jiayu,
and the Fuyang manuscript corpus (see Appendix 3:L).506 Three of these anecdotes (Xunzi,
Shuiyuan, Da Dai Liji) involve comparisons between water and a number of concepts or values
which include “virtue” (de 德), “propriety” (yi 義), and the “Way” (dao 道). Only one of these
versions—Mengzi 4B/18—parallels the language of Lunyu 9/17 in any way, and even there the
parallel consists of only four characters within a 59-character Mengzi quotation:
Xuzi said, “Zhongni often praised water, saying, ‘Water! Water!’ What did he glean from
water?” Mengzi said, “The water from springs gushes forth, stopping neither day nor night. It
flows onward only after filling the empty spaces, then spreads out to the four seas. Things with
a basis are like this, and this is what [Zhongni] gleaned. If something is without basis it is like
the summer rains that fill up the ditches but dry up while you stand there watching them. Thus
a noble man is ashamed if his reputation exceeds his character.”
徐子曰：仲尼亟稱於水，曰：水哉，水哉！何取於水也。孟子曰：原泉混混，不舍晝
夜，盈科而後進，放乎四海。有本者如是，是之取爾。苟為無本，七八月之間雨集，溝
澮皆盈；其涸也，可立而待也。故聲聞過情，君子恥之。
Given the prevalence of such episodes, Xuzi’s statement that Zhongni “often praised water” (qi
cheng yu shui 亟稱於水) might be read as an acknowledgement of the tradition’s popularity. In
asking “what Kongzi gleaned from water” (何取於水也), Xuzi seems to be soliciting a comment
506.Zhuangzi book 19, “Da sheng 達生” (SBCK 7/10b–11a), Shuiyuan book 17, “Za yan 雜言” (SBCK 17/16a–b),
and Kongzi jiayu book 8, “Zhi si 致思” (SBCK 2/9a–b), also include anecdotes which begin with Kongzi
“contemplating” (guan 觀) the “Lü cataract” (Lü liang 呂梁). However, they also have Kongzi interacting with a
mysteriously gifted swimmer who then instructs Kongzi in some lesson.
239
on an entire sub-genre of Kongzi saying, and Mengzi obliges with his own interpretation of what
water meant for Kongzi. Assuming that Mengzi 4B/18 predates Lunyu 9/17, it is tempting to
imagine that the Lunyu compilers selected the line “stopping neither day nor night” (bu she zhou
ye 不舍晝夜) for their own version because, like the Mengzi 4B/18 author, they were interested
in crafting an authoritative statement on an entire sub-genre of Kongzi saying.
If the Lunyu compilers did lift a line from Mengzi 4B/18, why that particular line? And what
of the other five characters (“Such is its passing!”)? Here the use of shi 逝 (to pass) is suggestive.
Outside of the Shi corpus, the word shi 逝 does not have a large presence in pre-Han and Western Han intellectual discourse.507 But it does form part of the stock poetic vocabulary of the Chuci collection,508 which includes a number of poems whose protagonists “contemplate” (guan 觀)
or “look down on” (lin 臨) water (see Appendix 3:M). If the Lunyu compiler was familiar with
the poetic tradition represented by the Chuci, that familiarity might have led him to select or
even compose the Kongzi saying at Lunyu 14/35 in response to the same poetic sensibility.
*
*
*
In section two I argued that the popularity of pre-Han proverb traditions exerted some pressure
on the Lunyu compiler to select Kongzi sayings on similar themes using similar language. The
analysis of this section suggests that popular Kongzi anecdotal traditions exerted a similar
pressure. Significantly, the Lunyu compilers seem not to have participated in these traditions
with their own fully developed versions of various Kongzi legends, instead preferring short,
507.Among ostensibly pre-Han texts, it occurs once in the Guanzi, twice in the Laozi, twice in the Lüshi chunqiui,
twice in the Mengzi (including one Shi quotation), twice in the Mozi, x23 in the Shijing, once in the Yanzi chunqiu,
once in the Zhanguo ce, x5 in the Zhuangzi, and once in the Zuozhuan (in a Shi quotation). Tellingly, eight of the
nine Shiji examples are found in “odes” (Shi 詩) and “performance texts” (fu 賦).
508.By my count, shi 逝 appears a total of 26 times in the Chuci.
240
standalone zi yue sayings over extended anecdotes. Their preference for epitomizing sayings may
also reveal something of their aspirations for the Lunyu as a text designed in part as an authoritative statement on pre-Lunyu Kongzi traditions. Lunyu 9/17, 9/28, and 15/2 may represent the
compilers’ effort to have the last word on early Kongzi legends, perhaps in order fix Kongzi’s biography against anti-Kongzi parodies like those found in the Mozi and Zhuangzi. Such sayings
would have satisfied an elite Western Han audience eager for a more predictable, and thus a
more quotable, Kongzi.
• Section #5: Kongzi on Kongzi
Both Lunyu 7/2 and 7/34 feature Kongzi describing himself as someone who simply “learns
without tiring and teaches others without wearying”:
Lunyu 7/2
The Master said, “Keep quiet and take note! I learn without tiring and teach others
without wearying. What [more] is there for me?
子曰：默而識之！學而不厭，誨人不倦，何有於我哉？
Lunyu 7/34 The Master said, “As for being sagely and humane, how would I dare? It might
perhaps be said that I conduct myself without tiring and teach others without
wearying—that is all.”
Gongxi Hua said, “This is precisely what your disciples cannot learn.”
子曰：若聖與仁，則吾豈敢？抑為之不厭，誨人不倦，則可謂云爾已矣。
公西華曰：正唯弟子不能學也。
As it happens, parallels in the Mengzi and Lüshi chunqiu frame the same core saying as Kongzi’s
response to a question from Zigong:
Mengzi 2A/2
Formerly Zigong asked Kongzi, “Master, are you a sage?” Kongzi said, “A
sage? I do not have the ability. I learn without tiring and teach without
wearying.”
昔者子貢問於孔子曰：夫子聖矣乎？孔子曰：聖則吾不能，我學不厭
而教不倦也。
241
Lüshi chunqiu 4/3, Zigong asked Kongzi, “How will later generations praise you, Master?”
“Zun shi 尊師”
Kongzi said, “What am I worthy to be praised for? Say nothing except that I
love learning without tiring and love teaching without wearying. That is all.
子貢問孔子曰：後世將何以稱夫子？孔子曰：吾何足以稱哉？勿已
者，則好學而不厭，好教而不倦，其惟此邪。
Despite the different introductory questions, with the Mengzi version asking about Kongzi’s
“sageliness” (sheng 聖 [*-eŋ]) and the Lüshi chunqiu about “praise” (cheng 稱 [*-əәŋ]) of Kongzi
generally, the parallels point to a shared tradition: first, Zigong asks Kongzi a flattering question;
next, Kongzi rejects the flattery (“As for being a sage, I do not have the ability”/“What am I worthy to be praised for?”); finally, he utters a saying which coordinates “learning” (xue 學) and
“not tiring” (bu yan 不厭) with “teaching” (jiao 教 or hui 誨) and “not wearying” (bu juan 不
倦). The Lüshi chunqiu version then concludes with an additional piece of modesty (“That is
all”).509
The appearance of Zigong in these dialogues is probably not a coincidence. Elsewhere in the
Lunyu, Lunyu 19/22–25 feature Zigong’s descriptions of his master, 15/24 has Zigong eliciting
the “one saying that can be practiced to the end of one’s life,” and 14/28 has him commenting on
a Kongzi saying with the words, “The Master describes himself” (夫子自道也). Outside the text,
it is Zigong who criticizes Duke Ai’s 哀公 eulogy of Kongzi in the Zuozhuan (see p. 99) and
who elicits Kongzi’s description of himself as a “homeless dog” (sang jia gou 喪家狗) in the
Hanshi waizhuan, “Kongzi shijia,” Baihu tongyi, Kongzi jiayu, and Lunheng.510 As we saw in
509.A parallel from the “Kongzi shijia” (Shiji 47.1928) also includes this Kongzi saying directed at Zilu: “Kongzi
learned of this and said, ‘You [=Zilu], why didn’t you say, “As a man he learns the Way without tiring and teaches
others without wearying. When exercised he forgets to eat, when joyful he forgets his worries, and he does not know
the onset of old age” (孔子聞之，曰：由，爾何不對曰：其為人也，學道不倦，誨人不厭，發憤忘食，樂以
忘憂，不知老之將至云爾。). Curiously, this saying seems to be a mash-up of Lunyu 7/2 and 7/19.
510.Hanshi waizhuan 9 (SBCK 9/10a); Shiji 47.1921; Baihu tongyi 31, “Shou ming 壽命” (SBCK 8/5a); Kongzi
jiayu 22, “Kun shi 困誓” (SBCK 5/23a): Lunheng 11, “Gu xiang 骨相” (SBCK 3/10b). In the “Wei jiangjun Wenzi
衛將軍文子” (“Wenzi, General of Wei”) chapter of the Da Dai Liji, Zigong is also asked to evaluate various
242
chapter one (p. 127), Zigong was also the disciple most often asked to characterize Kongzi’s
virtues. There I argued that Zigong’s reputation as the most rhetorically gifted disciple allows us
to read such episodes as statements on the ineffability of Kongzi’s virtue. Zigong’s symbolic status probably explains his involvement in episodes in which Kongzi is asked to describe himself,
his presence marking that a saying dealt with the problem of matching language to reality.
Here it might be objected that my translation of the first four characters of Lunyu 7/2—“Keep
quiet and take note” (默而識之)—goes against a longstanding commentarial tradition which
reads them as part of Kongzi’s self-description:
<Xing Bing 邢昺 (931–1010)>In this paragraph Zhongni says that he records without
speaking, that he learns from the ancients without tiring his heart, and that he teaches others
without growing weary (此章仲尼言已不言而記識之，學古而心不厭，教誨於人不有倦
息。).511
<James Legge’s (1815–1897) translation>The silent treasuring up of knowledge; learning
without satiety; and instructing others without being wearied—which one of these things
belongs to me?512
<D.C. Lau translation>Quietly to store up knowledge in my mind, to learn without flagging, to
teach without growing weary. For me there is nothing to these things.513
<Edward Slingerland translation>Remaining silent and yet comprehending, learning and yet
never becoming tired, encouraging others and never growing weary—these are tasks that
present me with no difficulty.514
Commentators from Li Chong 李充 (fl. fourth century) onward have likewise described Lunyu
7/2 as a statement on the “three-fold path” (san hang 三行) on the assumption that Kongzi’s selfdescription consisted of three distinct elements.515
Kongzi disciples.
511.Lunyu zhengyi 7/15 (Shisanjing zhushu p. 2481).
512.Legge 1867, p. 153
513.Lau 2000, p. 57.
514.Slingerland 2003, p. 64.
515.Lunyu jishi pp. 436–39.
243
But there are a number of problems with this reading, beginning with the fact that the first of
the three paths is neither semantically nor grammatically parallel to the other two. If the “three
paths” really were intended to be read as a group, we might expect the first path to parallel the
second just as the second parallels the third (e.g., with the line “Know without speaking” *識而
不言). Second, the notion of “learning with tiring and teaching without wearying” appears in a
number of Eastern Han texts but never alongside mo er shi zhi 默而識之. If Lunyu 7/2 was originally conceived of as a “three-fold path,” why did all of the Han authors who quoted Lunyu 7/2
fail to notice that fact? Third, as discussed in chapter one (p. 51), a number of early texts feature
Kongzi ordering his disciples to “take note of” (zhi 志 or shi 識) his instructions. Zhuangzi book
21, “Tian Zifang 田子方,” even has Kongzi telling his most cherished disciple to mo 默 (“be quiet”), the same word which opens Lunyu 7/2:
Yan Yuan asked Zhongni, “Did King Wen fall short? Why did he use a [fake] dream to get his
way?” Zhongni said, “Quiet! Do not speak. King Wen was perfect, so how can you evaluate
and criticize him? He only used [the dream] to handle that particular situation.”
顏淵問於仲尼曰：文王其猶未邪？又何以夢為乎？仲尼曰：默！汝無言。夫文王盡之
也，而又何論刺焉？彼直以循斯須也。516
Translating Lunyu 7/2 as “Keep quiet and take note” in parallel with this Zhuangzi dialogue has
the additional advantage of highlighting a convergence with the Lüshi chunqiu parallel, in which
Kongzi prefaces his instruction with a bit of scolding: “Enough!” (勿已者). Thus, the evidence
from Lunyu parallels reveals Lunyu 7/2, like 2/17, as a saying extracted from an earlier dialogue,
one that still bears the scars of its extraction.517
516.Zhuangzi 21 (SBCK 7/38a–b; KZJY 17.15/p. 392). Also note Kongzi’s contempt for Zigong at, e.g., Xunzi 29,
“Zi dao 子道” (SBCK 20/10a–b; KZJY 2.12/p. 18) (Kongzi said, “Ci, you are a petty man who doesn’t understand”
孔子曰：小人哉！賜不識也。), and Hanshi waizhuan 4 (SBCK 4/6a–b; KZJY 7.10/p. 149) (Kongzi said, “...Ci,
you are a sorry envoy. How can you possibly understand ritual?” 孔子曰：...賜，寡使也，何足以識禮也！).
517.The opening of Lunyu 7/34 (“As for being sagely and humane, how would I dare?”) probably also points to an
earlier dialogical context.
244
But the question remains: why would the Lunyu compilers include multiple versions of the
same Kongzi saying within a single Lunyu book? Part of the answer to that question lies in the
observation that “learn without tiring and teach without wearying” is one of the very few Lunyu
Kongzi sayings with multiple parallels in pre-Han texts. Like the various Lunyu sayings connected to Chen and Cai stories, the inclusion of both Lunyu 7/2 and Lunyu 7/34 might simply reflect
the original saying’s popularity in the period. More significantly, the clustering of Kongzi selfappraisals in Lunyu book 7 speaks the Lunyu compilers’ evident interest in these sayings as selfappraisals. Who better to fix an authoritative image of Kongzi than Kongzi himself?
• Section #6: Lunyu 12/1 & 12/2 and the situated saying
In the examples discussed so far I have tried to establish that the Lunyu compilers engaged earlier material with an eye to selecting memorable sayings. Once selected, evidence of sayings’ preLunyu history was (mostly) cast aside. However, the Lunyu does not consist solely of isolated zi
yue sayings. Consider Lunyu 12/1, whose likeliest precursor is a Kongzi comment from the
Zuozhuan entry for 530 BCE, the twelfth year of Duke Zhao’s 昭公 reign (see Appendix 3:N).
The focal point in both the Lunyu and Zuozhuan versions is the idea that “humaneness” (ren 仁)
consists of “subduing oneself and reviving ritual” (ke ji fu li 克己復禮).518 Whereas the
Zuozhuan saying introduces that idea as part of a comment on the story of King Ling of Chu 楚
靈王, the Lunyu casts it as a response to a question from Yan Yuan about “humaneness” (ren
仁). The Zuozhuan Kongzi also attributes the core saying to an “ancient maxim” (gu ye you zhi
518.For a parallel discussion see Haupt 2006, pp. 126–127
245
古也有志) which is not mentioned by the Lunyu Kongzi.
The core saying (“subduing oneself and reviving ritual”) aside, Lunyu 12/1 has two additional parallels in the Zuozhuan. The first is the line “One becomes humane from oneself—how
could it be from another?” (為仁由己，而由人乎哉？), whose core template (X由己，由人乎
哉) appears twice in unattributed judgments in the Zuozhuan but nowhere else in the extant literature that I am aware of.519 In these comments the anonymous Zuozhuan narrator steps out of the
narrative to cast judgment on various historical actors and draw generally applicable morals,
much like the Kongzi comments on anecdotes discussed in chapter one (pp. 48–53). The second
parallel appears not in Lunyu 12/1 but in Lunyu 12/2, which also features Kongzi responding to a
question from a disciple about “humaneness” (ren 仁) with a saying whose earliest parallel is in
the Zuozhuan (Appendix 3:O). Like 12/1, Lunyu 12/2 also ends with a disciple professing his
“lack of intelligence” (bu min 不敏) and “begging to put this saying into practice” (qing shi si yu
yi 請事斯語矣).520 The clustering of Zuozhuan parallels in Lunyu 12/1 and 12/2 is compelling
evidence for reading both episodes as actual adaptations from the Zuozhuan.
If so, how were they adapted, and what does the form of the adaptations say about the interests of the Lunyu compilers? One way to approach this question is to ask why Lunyu 12/1 was
preferred over dozens of other Kongzi sayings in the Zuozhuan. Judging from the opening lines
of Lunyu 12/1 and 12/2—“So-and-so asked about humaneness” (wen ren 聞仁)—the answer to
that question has something to do with the compilers’ interest in “humaneness” (ren 仁) as an ab-
519.See Zuozhuan Xi 20 (SBCK 6/4b) (“Success and failure comes from oneself; how could they come from
another?” 善敗由己，而由人乎哉。) and Zuozhuan Cheng 6 (SBCK 12/13b) (“One establishes a martial [virtue]
through himself; it is not from another” 立武由己，非由人也。).
520.This is a generic element of a number of Kongzi dialogues. Variants of the formula appear six times in the
Hanshi waizhuan alone, twice in the mouths of Kongzi’s interlocutors Ji Kangzi 季康子 (SBCK 3/14a) and Zigong
(SBCK 7/15a).
246
stract concept. The word ren 仁 appears in only three Zuozhuan Kongzi sayings, two of which
are dependent comments on specific historical individuals, Zichan 子產 and Zang Wenzhong 臧
文仲.521 The third is the parallel to Lunyu 12/1: “subduing oneself and reviving ritual is humaneness.” If the Lunyu compilers were scouring Zuozhuan Kongzi sayings looking for an abstract,
universally applicable statement on ren, “subduing oneself and reviving ritual” would have been
their only option.
Lunyu 12/2 is not attributed to Kongzi in the Zuozhuan. However, it may have caught the
compilers’ eye because it was an independent saying (prefaced by the quotation marker “your
subject has heard” [chen wen 臣聞) which made a strong claim about “the standard of humaneness” (ren zhi ze ye 仁之則也). Although ren is an oft-invoked concept in the Zuozhuan, particularly in speeches, this is the only other line I have found in the text which offers anything like an
explicit definition of ren.522
Given their preference for extracted sayings elsewhere in the Lunyu, why did the Lunyu compilers decide to contextualize these particular sayings as conversations between Kongzi and his
disciples? The form of the dialogue in Lunyu 12/1, which amplifies each element of the original
saying in three successive stages, suggests one possible answer:
Subduing oneself
and reviving ritual
a. 克己
b. 復禮
is humaneness.
c. 為仁
↓
c1. If in one day one can subdue himself
and revive ritual then the entire world
will return to humaneness.
一日克己復禮，天下歸仁焉。
↓
521.For these comments, see Zuozhuan Wen 2 (SBCK 8/5a; KZJY p. 413) and Zhao 12 (SBCK 22/15b; KZJY pp.
417–18)
522.Ren 仁 is glossed in several passages in the Zuozhuan alongside other values such as xin 信 and yi 義. But the
Zuozhuan parallels of Lunyu 12/1 and 12/2 consider ren 仁 independently of other values.
247
a1. One becomes humane from himself—
how could it be from another?
為仁由己，而由人乎哉？
↓
b1. If it is not ritually proper then do not look at it;
if it is not ritually proper then do not listen to it;
if it is not ritually proper then do not speak it;
if it is not ritually proper then do not act on it.
非禮勿視，非禮勿聽，非禮勿言，非禮勿動。
The first part (c1) expands the scope of the original saying by emphasizing its potency: an individual (presumably a ruler) who “subdues himself and revives ritual” even for a single day will
cause “all under heaven” (tian xia 天下) to be humane. Here we might also note a phonological
resonance between the Lunyu’s gui 歸 (*kwəәj) and the Zuozhuan’s wei 為 (*ɢwaj). The next line
(a1) amplifies the notion of “subduing oneself” (ke ji 克己), and Yan Yuan’s follow-up question
(“May I ask about the particulars”) prompts a response which loosely parallels a Kongzi saying
from the Shanghai Museum “Junzi wei li 君子為禮” manuscript (see Appendix 2:V, parallels 4
& 5), although in that text the focus is yi 義 (propriety) and not li 禮.
In short, the lun-ing of Lunyu 12/1 (and perhaps also 12/2)523 took place in several steps.
First, the Lunyu compilers might have scanned the Zuozhuan for maxims on ren 仁 as an abstract
concept. Not content to let those sayings speak for themselves, probably because the Zuozhuan
did not attribute them to Kongzi, they contextualized them as dialogues between Kongzi and his
disciples.524 Repurposing the dependent Zhongni yue comment as a scene of instruction and eliding the words “there is an ancient maxim” bolstered the impression that the core sayings orig-
523.While the structure of Lunyu 12/2 is less neat, the line “you will incur resentment neither in your state nor in
your home” (在邦無怨，在家無怨。) seems to amplify the core saying’s contrast between “being away from
home” (chu men 出門) and “directing the people” (shi min 使民) in the course of one’s official duties.
524.Christiane Haupt (2006, p. 50–52) has noted that the Kongzi jiayu parallels of many Zuozhuan Kongzi sayings
are also presented as dialogues which emphasize Kongzi’s role as a master. By contrast, the Kongzi comments in the
Zuozhuan are never prompted by or directed at an interlocutor.
248
inated with Kongzi and Kongzi alone. Perhaps because they especially valued the core sayings of
Lunyu 12/1 and 12/2, the compilers then elaborated on them before concluding with the disciples’ words, “I beg to put this saying into practice.” Far from representing an actual scene of instruction, the dialogical framework of Lunyu 12/1 and 12/2 was simply a way to assert Kongzi’s
ownership of the sayings, and to emphasize their value.
• Section #7: Lunyu 7/1 and the appropriated Kongzi saying
Any list of the most famous and influential Lunyu Kongzi sayings would have to include Lunyu
7/1:
The Master said: “Insofar as I transmit without originating and trust in and love the ancient, I
would liken myself to our Old Peng.”
子曰：述而不作，信而好古，竊比於我老彭。525
In addition to serving as a slogan for authors throughout the Chinese tradition, Lunyu 7/1 had a
profound influence on the imagination of Kongzi in later periods. To take just one example, Zhu
Xi 朱熹 (1130–1200), the figure ultimately responsible for recanonizing the Lunyu as one of the
Four Books (Si shu 四書), had this to say about Lunyu 7/1 in his Si shu zhangju jizhu 四書章句
集注 (Collected Chapter-and-Verse Commentaries on the Four Books).
Kongzi edited the Shi and Shu, he fixed the Li and Yue, he enlarged the Zhouyi, and he
emended the Chunqiu. In every instance he transmitted the old institutions of the former kings
without ever inserting something original, thus he spoke of himself in this way. But it was not
just that he did not dare to assume the role of an originating sage. He also did not dare to
conspicuously include himself among the ancient men of accomplishment. The more
flourishing his virtue the more humble he was in his heart, so that he himself was unaware of
525.The identity of “Old Peng,” or perhaps “Lao and Peng,” is a question that has greatly exercised traditional
commentators. Here I follow Bao Xian’s suggestion that lao Peng 老彭 refers to Pengzu 彭祖 (SBCK 4/1a),
apparently a sagely minister under the Shang 商. As mentioned earlier, given other resonances between the Lunyu
and Chuci collection, I would not rule out the possibility that the Peng 彭 of Lunyu 7/1 is the Peng Xian 彭咸
referenced in the coda of the “Li sao 離騷” (“Encountering Sorrow”): “I shall go and join Peng Xian in the place
where he abides” (吾將從彭咸之所居). For that translation, see Hawkes 1985, p. 78.
249
how self-deprecating his words were. In his time originators were in short supply, and so the
Master collected the great accomplishments of various sages and harmonized them together.
Although the content [of Kongzi’s teachings] was “transmitted” (shu 述), his achievement was
twice that of “originating” (zuo 作). This, too, cannot be ignored.
孔子刪詩，書，定禮、樂，讚周易，修春秋，皆傳先王之舊，而未嚐有所作也，故其自
言如此。蓋不惟不敢當作者之聖，而亦不敢顯然自附於古之賢人；蓋其德愈盛而心愈
下，不自知其辭之謙也。然當是時，作者略備，夫子蓋集群聖之大成而折衷之。其事雖
述，而功則倍於作矣，此又不可不知也。526
Although Lunyu 7/1 is sixteen characters long, what really interested Zhu Xi was the four-character phrase “transmitting without originating” (shu er bu zuo 述而不作) with its implied distinction between the “originating” (zuo 作) teachings of the sages and the humble “transmissions” (shu 述) of lesser men. For Zhu Xi, Kongzi’s “self-deprecation” (qian 謙) presented a
problem: what is one to do when an infallible sage denies his own sageliness? His solution was
two-fold: on the one hand, he argued that Kongzi was so supremely virtuous and humble that he
underestimated his own greatness; on the other, Kongzi’s textual activities represented a unique
kind of “transmission” which transcended common authorship and facilitated the creation of the
Five Classics.
What did Lunyu 7/1 mean to authors in the early period? The Lunyu jijie 論語集解 (Collected Explications of the Lunyu) preserves a line of commentary on Lunyu 7/1 which it attributes to
Bao Xian 包咸 (8 BCE–65 CE), one of the Lunyu transmitters named in the “Yiwen zhi;” however, that commentary focuses on the final six characters without even mentioning the first
four.527 To appreciate the significance of shu er bu zuo 述而不作 in the early period, we must
turn to the evidence from textual parallels. Not all of the examples listed in Appendix 3:P can be
read as “parallels” in the sense described in the introduction to chapter two, i.e., as possible bor-
526.Sishu zhangju p. 93.
527.SBCK 4/1a.
250
rowings from a Lunyu text. Here I have included any and all passages in which zuo 作 and shu
述 (and its variants)528 are related in a way which at least vaguely resembles the first four characters of Lunyu 7/1.
Clearly, expressions coordinating shu 述 and zuo 作 appear in a wide range of texts from the
Warring States through the Han period. These expressions fall into three groups: passages in
which the original creations of the supremely virtuous “sages” (shengren 聖人) or “great men”
(daren 大人) are contrasted with the transmissions of less virtuous “men of accomplishment”
(xianren 賢人) or “followers” (dizi 弟子) (#4, #5, #6, #8, #9, #17); those in which the generic
junzi or true king is enjoined to transmit without originating (#1, #2, #3, #7, #10, #11, #13, #15);
and those like Lunyu 7/1 in which Kongzi self-applies the phrase to claim that he merely transmitted without originating (#12, #14, #18, #19). Not all of these parallels are independent sayings; some simply integrate the shu 述/zuo 作 dichotomy into longer discourses.
According to the standard view of the Lunyu, most if not all of these passages can be interpreted as parallels or allusions to Lunyu 7/1. Thus, in a comment on example #2, Wu Yujiang 吳
毓江 in the Mozi jiaozhu 墨子校注 (The Mozi, Collated and Annotated) simply wrote, “The
‘Shu er’ chapter of the Lunyu says, ‘Transmit without originating’” (論語述而篇曰：述而不
作。), presumably because he believed that the Mozi was responding to the Lunyu.529 But what is
remarkable about this list is how few parallels actually cite Kongzi or the Lunyu as the source.530
528.In example #2, shu 術 (*Cəә-lut) is a homonym for shu 述 (*Cəә-lut) while xun 循 (*səә.lun) in examples #1, #3,
and #4 appears to be a loosely rhymed semantic variant.
529.The first Mozi example can be found at Mozi jiaozhu p. 437 & p. 449, n. 61, the second at Mozi jiaozhu p. 660
& p. 680, n. 96.
530.While the “Zhong yong 中庸” chapter of the Liji (example #5) prefaces a significant variant (“A father
originates, a son transmits”) with zi yue, and the Mozi (example #1) attributes a close variant (“A noble man follows
without originating”) to the Ru 儒 collectively, the former is a significant variant of Lunyu 7/1 and the latter does not
mention Kongzi.
251
Of the eighteen passages that coordinate shu and zuo in a manner reminiscent of Lunyu 7/1, only
three (#14, #18, #19) explicitly attribute shu er bu zuo 述而不作 to Kongzi or the Lunyu, and all
three treat shu er bu zuo as Kongzi’s self-description rather than as a general prescription. One
more passage (#12) quotes the first four characters of Lunyu 7/1 together with the second four
characters, a sign that its author was familiar with Lunyu 7/1. However, the earliest of these attributions appears in Ban Gu’s (32–92 CE) Hanshu, a text compiled well into the Eastern Han period.531 Also telling is the fact that the phrase xin er hao gu 信而好古 (“trust in and love the ancient”) does not appear in the extant literature until the Eastern Han period, and then only in
conjunction with shu er bu zuo. This suggests that the phrase was original to the Lunyu.
In light of the developments in Kongzi quotation practice outlined in chapter two, it would
seem that the association of shu er bu zuo with Kongzi was not fixed in the popular imagination
until well after the Lunyu had attained its status as the pre-eminent source of Kongzi sayings.
The fact that the only early texts to cite Kongzi as the source of “transmit without originating”
quote the Lunyu version word-for-word further suggests that the saying’s association with
Kongzi derived solely from the Lunyu and not from some other source. This micro analysis of
Lunyu 7/1 thus corroborates the macro analysis of Kongzi quotation practice in chapter two: as
Han authors began relying primarily on the Lunyu for their Kongzi sayings, individual sayings
with a history beyond the Lunyu came to be seen strictly as Lunyu Kongzi sayings, as a result of
which non-Lunyu variants and associations were cast aside. However, it is also important to note
that even in the Eastern Han, Wang Chong in a treatise on authorship (#17) in which he else-
531.The Hanshu example appears in the introduction to the “Ru lin 儒林” (“Forest of Ru”) chapter and is thus likely
to be part of the latest layer of the text. Although it does not quote Kongzi explicitly, its context—a brief biography
of Kongzi—removes any uncertainty about the intended source of the attribution.
252
where praises Kongzi as a model author could quote a variant of shu er bu zuo without mentioning either Kongzi or the Lunyu.
Why would the Lunyu compilers select shu er bu zuo for inclusion if it did not already have a
history as a Kongzi saying? The frequent appearance of the shu 述/zuo 作 dichotomy in the early
literature suggests an answer: by stamping Kongzi yue onto a saying that was authoritative by
virtue of being widely recognizable, the Lunyu compilers presented Kongzi as the original source
of that wisdom, much as other Lunyu sayings appropriated bits of received wisdom more commonly associated with other masters. Lunyu 7/1 thus points to another recipe for the construction
of a Lunyu Kongzi saying: (1) select a saying that is generically authoritative by virtue of its circulation and recognizability; (2) attribute it to “The Master”; (3) if the resulting saying is too
brief, add a parallel line—here “trust in and love the ancient” (xin er hao gu 信而好古), which
rhymes gu 古 (*kʕaʔ) with zuo 作 (*tsʕak)—for the sake of balance;532 and (4) personalize the
saying (wo 我) in order to cement its association with Kongzi.
step 1: selection
transmit without originating
述而不作
step 2: attribution
The Master said, “Transmit without originating.”
子曰：述而不作。
step 3: amplification
The Master said, “Transmit without originating,
trust in and love the ancient.”
子曰：述而不作，信而好古。
step 4: personalization The Master said, “Insofar as I transmit without originating and
trust in and love the ancient, I would liken myself to our Old Peng.”
子曰：述而不作，信而好古，竊比於我老彭。
Here it is important to keep in mind that Lunyu 7/1, like the Lunyu generally, makes no claim
whatsoever about the ultimate source of the Kongzi saying. It is entirely possible that the Lunyu
532.Sayings of only four characters are exceedingly rare in the Lunyu and elsewhere, one notable exception being
Lunyu 15/39.
253
compilers and their intended audience would have understood “transmit without originating” as a
generic proverb, in which case Lunyu 7/1 might have read as an implicit quotation: “The Master
said, ‘Insofar as I “transmit without originating...”’” Not until later did authors come to think of
Lunyu 7/1 as belonging to Kongzi and Kongzi alone.533
*
*
*
Lunyu 7/1 is by no means the only Lunyu Kongzi saying whose parallels indicate that it circulated originally as a generic saying. In chapter two I noted that parallels between Lunyu Kongzi
sayings and non-Kongzi material far outnumber parallels with early Kongzi material, particularly
in pre-Han and early Han texts (see Appendix 2:X). To name just one example, sayings built on
the idea that a junzi “reforms” (gai 改) himself when he commits “errors” (guo 過) or is “deficient” (bu shan 不善) appear six times in the Lunyu (1/8, 7/3, 7/22, 9/25, 15/30, and 19/21) as
well as in numerous early texts (see Appendix 3:Q). In none of these instances is the saying
attributed to Kongzi. After the Lunyu, the earliest text to credit Kongzi with this cliché is the
“Kongzi shijia.” And of the six versions of this idea in the Lunyu, Lunyu 7/3 most closely follows the 7/1 template insofar as it incorporates a widely circulating proverb into a longer saying
sandwiched between “The Master says” and a personalizing statement (“these are my worries”
是吾憂也).
*
*
*
From the perspective of their parallels, Lunyu 7/1 and these other examples cast the Lunyu compilers as collectors of clichés. They identified widely circulating proverbs which they then
stamped with Kongzi’s imprimatur, a strategy which might have enhanced the text’s appeal for
533.Kimura (1971 pp. 251–252; 320–321) also notes the phenomenon whereby common clichés were adapted into
Kongzi sayings.
254
an elite Western Han audience by giving Lunyu Kongzi sayings an air of familiarity. “You know
those proverbs that you are so fond of repeating?” the Lunyu compilers in effect said to their audience, “Kongzi said them first.” As Aristotle observed in the Rhetoric,
Maxims make one great contribution to speeches because of the uncultivated mind of the
audience; for people are pleased if someone in a general observation hits upon opinions that
they themselves have about a particular instance...Thus, one should guess what sort of
assumptions people have and then speak in general terms consistent with these views.534
Whether or not the Lunyu compilers shared Aristotle’s disdain for ancient audiences, their appropriation of previously unappropriated sayings is one sign that they might have shared his view of
the rhetorical value of popular proverbs.
• Section #8: Template sayings
In The Original Analects, Taeko and Bruce Brooks had this to say about one of the core themes
of the earliest stratum of the Lunyu:
Another type of tension which can be seen in the early Analects layers is that between the old
warrior elite, who had once monopolized most court positions, whose social ideal was the
junzi 君子 or “gentleman,” the code of those who were born to rule others, and what the text
calls the xiaoren 小人 or “little people,” newcomers of artisan or entrepreneurial origin, with
the value system of those who survive by wit rather than force: know-how, personal charm,
ingratiating speech, avoiding rather than facing danger, and a keen eye for the bottom line.535
Given that the the junzi/xiaoren dichotomy appears nineteen times in eighteen different Lunyu
entries, it would seem that this “tension” was indeed on the mind of the Lunyu compilers (Appendix 3:R). Extending the same logic to other early texts, the dichotomy’s prevalence throughout the early literature would indicate that this social division preoccupied a majority of early authors. By my count, the junzi/xiaoren contrast appears four times in the Da Dai Liji, x7 in the
534.Rhetoric 1395b; trans. Kennedy 1991, p. 186.
535.Brooks & Brooks 1997, p. 11. See also Slingerland’s (2003, p. 211) comment on Lunyu 17/23: “’Gentleman’
and ‘common person’ are here meant in terms of social rank.”
255
Guanzi, x4 in the Guodian corpus, x6 in the Guoyu, x8 in the Han Feizi, x3 in the Hanshi
waizhuan, x19 in the Huainanzi, x20 in the Liji, x4 in the Lüshi chunqiu, x5 in the Mengzi, x3 in
the Mozi, x2 in the Shangjun shu, x3 in the Shangshu, x2 in the Shangshu dazhuan, x12 in the
Shiji, x4 in the Shijing, x26 in the Shuiyuan, x16 in the Wenzi, x1 in the Xinshu, x2 in the Xinyu,
x46 in the Xunzi, x6 in the Yanzi chunqiu, x1 in the Yi Zhoushu, x1 in the Zhanguo ce, x18 in the
Zhouyi, x7 in the Zhuangzi, and x15 in the Zuozhuan.
But there is another way of interpreting junzi/xiaoren statements. In fact, their prevalence in
the early literature is itself an argument against taking it as evidence of a real conflict between
actual junzi and actual xiaoren. In a number of instances, the “tension” identified by the Brooks
seems rather lukewarm. Consider Lunyu 12/16:
The Master said, “A junzi makes others realize their virtues; he does not make them realize
their vices. A xiaoren is the opposite of this.”
子曰：君子成人之美，不成人之惡。小人反是。
The author of this saying did not even bother to spell out the ways in which xiaoren are distinct
from junzi (e.g., with the line “a xiaoren makes other realize their vices, he does not make them
realize their virtues” *小人成人之惡，不成人之美). He simply states that “a xiaoren is the opposite of this,” presumably because he knew that his audience would know that xiaoren and junzi
are opposites by default.536 Several other Lunyu sayings (2/14, 13/23, 13/26, 14/23, 15/21) follow
the formula “A junzi is X and a xiaoren is not X” in similarly rote fashion. More often than not, a
xiaoren statement which follows a junzi statement (or vice versa) is entirely derivative of the other and contributes nothing new to the whole.
If so, then why was the junzi/xiaoren dichotomy so frequently invoked? I would argue that it
536.Xunzi book 3, “Bu gou 不苟” (SBCK 2/4a), puts this point succinctly: “The noble man is the petty man’s
opposite” (君子小人之反也。).
256
is an early example of what Stephen Owen has referred to in another context as the “discourse
machine,” the set of informal rules which dictate how discrete ideas are amplified and embedded
within a discourse.537 Put simply, the junzi/xiaoren binary is a rhetorical trope whose use implies
nothing about the intellectual or social milieu of the Lunyu compilers. These sayings are not
about junzi and xiaoren. They are an artifact of early rhetorical practice, a way of expressing relationships between concepts and perhaps also of balancing and extending sayings that would
otherwise feel too brief. Take the example of Lunyu 13/23 (“A noble man harmonizes without
agreeing, a petty man agrees without harmonizing” [君子和而不同，小人同而不和。]), which
features two roughly synonymous terms—he 和 (harmonize or be harmonious) and tong 同 (be
the same as or agree with). But there is a key difference between them according to the Lunyu:
he 和 is morally praiseworthy, tong 同 morally blameworthy. (There may also be an implication
that “harmonizing” is superior to thoughtless agreement, as Yanzi 晏子 argues in an episode
from the Zuozhuan. However, the Lunyu version does not develop this connection.538) Associating he 和 with the junzi and tong 同 with the xiaoren was simply a convenient way of clarifying
the terms’ moral valence.
One sign that early authors also thought of the junzi/xiaoren binary as a rhetorical device and
537.Owen 2001, p. 175: “In this essay I would like to consider Liu Xie’s arguments not as the ‘expression’ of ideas
already fully formed and fixed, but rather as a process of exposition. This process is not unitary: In many cases we
can identify two ‘players’ contending for control of the exposition. One of these players we will call ‘Liu Xie,’ a
human character with beliefs, an education of received ideas, and common sense. The other main player is the
rhetoric of parallel exposition, what I call the ‘discourse machine,’ which produces utterances by its own rules and
requirements. Although Liu Xie would like to believe that these two players are a perfect unity and although modern
criticism treats them as such, much in Wenxin diaolong becomes clearer if we understand the text as dialogic. We
often see the productive rhetoric of the discourse machine processing an initial statement and amplifying it
according to predictable rules. No less often we see Liu Xie following in the traces of the discourse machine,
ingeniously correcting unwanted utterances and attempting to make the argument conform to beliefs, received ideas,
and common sense.”
538.Zuozhuan Zhao 20 (SBCK 24/11a). Slingerland 2003, pp. 149–50, notes this connection and also includes a
translation of the Zuozhuan episode.
257
not as a description of some social tension is the scarcity of parallels among junzi/xiaoren statements in the early corpus. The only early parallels to Lunyu 13/23 I have found appear well into
the Eastern Han, and this pattern holds for other junzi/xiaoren sayings as well. Despite their
prevalence in the early literature, with a handful exceptions authors tended not to borrow others’
junzi/xiaoren statements. Instead, they used the junzi/xiaoren template to create new sayings and
discourses for their own purposes.
*
*
*
Of the many phrases and formulae one finds repeated throughout the Lunyu, the contrast between
the world or a state with the Way (tianxia you dao 天下有道 or bang you dao 邦有道) and one
without (tianxia wu dao 天下無道 or bang wu dao 邦無道) is among the most frequently invoked, occurring ten times in seven sayings across five different Lunyu books (Appendix 3:S).
Here again we might be tempted to conclude that the contrast between polities with the Way and
those without was somehow central to the thinking of the Lunyu compilers. But in what sense?
There is some overlap among these sayings, with Lunyu 8/13 and 14/1 stating that earning a
salary in a state without the Way is “shameful” (chi 恥) and Lunyu 8/13 and 15/7 expressing the
idea that one should serve a state with the Way and refuse to serve one without. However, the
other instances do not seem to cohere around a single theme.
The Lunyu was certainly not the only early text to make use of the “polity with the Way/polity without the Way” dichotomy (see Appendix 3:T), which has at least twenty-seven instances in
fifteen texts outside the Lunyu. A handful of these parallels (#7, #13, 21, #22, #23) echo Lunyu
8/13 and 15/7 insofar as they advocate serving a polity with the Way and retiring from one without. But most instances have little to do with the Lunyu or even with one another. As with the
junzi/xiaoren dichotomy, it would seem that early authors applied the formula when they wanted
258
express approval for one activity and disapproval for its opposite, as when the author of the Shiji
monograph on calendrics (#14) invoked it to stress the importance of calendar-keeping to a wellgoverned state.
*
*
*
Another template emerges from a series of Kongzi sayings at Lunyu 4/2, 6/23, 9/29, 14/4, and
14/28 (Appendix 3:U). Contrastive dyads and triads involving “wise men” (zhi zhe 知者), “humane men” (ren zhe 仁者), “brave men” (yong zhe 勇者) and other types likewise appear in a
number of early texts (Appendix 3:V).
Xunzi book 29, “Zi dao 子道” (“The Way of Sons”), provides some evidence that early authors also thought of zhi zhe 知者/ren zhe 仁者/yong zhe 勇者 contrasts as rhetorical templates:
When Zilu entered the Master said, “Zilu, what is a wise man like? What is a humane man
like?” Zilu answered, “A wise man makes others know him and a humane man makes others
care for him.” The Master said, “You can be called a ‘man of service.’” When Zigong entered
the Master said, “Zigong, what is a wise man like? What is a humane man like?” Zigong
answered, “A wise man knows others and a humane man cares for others.” The Master said,
“You can be called a ‘noble man of service.’” When Yan Yuan came in the Master said, “Yan
Hui, what is a wise man like? What is a humane man like?” Yan Yuan answered, “A wise man
knows himself and a humane man cares for himself.” The Master said, “You can be called an
‘enlightened noble man.’”
子路入，子曰：由！知者若何？仁者若何？子路對曰：知者使人知己，仁者使人愛己。
子曰：可謂士矣。子貢入，子曰：賜！知者若何？仁者若何？子貢對曰：知者知人，仁
者愛人。子曰：可謂士君子矣。顏淵入，子曰：回！知者若何？仁者若何？顏淵對曰：
知者自知，仁者自愛。子曰：可謂明君子矣。539
The episode resembles the interview scenes discussed in section one insofar as they feature
Kongzi asking questions of his disciples and then using their responses to judge their character.
In light of the argument of this section, we might also read it as a rhetorical exercise in the use of
the zhi zhe/ren zhe template. Kongzi defines the template with his initial question—“What is a
wise man like? What is a humane man like?”—and his disciples fill in the gaps with three dis-
539.Xunzi 29 (SBCK 20/13a–b; KZJY 3.19/pp. 33–34).
259
tinct versions. Although Kongzi judges Yan Yuan’s version to be the best, to the anecdote’s audience the content of these sayings might have mattered less than understanding the template itself and how to apply it to new contexts.
*
*
*
Why were the Lunyu compilers so interested in “template” sayings? The most striking feature of
the Lunyu’s use of the “noble man/petty man,” “with the Way/without the Way,” and “wise man/
humane man” dichotomies is how prevalent they are in the Lunyu compared to other early
texts.540 Generally speaking, other texts do not present template sayings in isolation but instead
built them into longer discourses. A few Lunyu passages even combine template sayings in a way
that calls attention to the template, e.g., in Lunyu 4/11, 6/23, and 8/13:
Lunyu 4/11 The Master said, “A junzi cherishes virtue, a xiaoren man cherishes land; a junzi
cherishes punishments, a xiaoren cherishes mercy.”
子曰：君子懷德，小人懷土；君子懷刑，小人懷惠。
Lunyu 6/23 The Master said, “A wise man delights in water, a humane man delights in mountains. A wise man is in motion, a humane man is at rest. A wise man is joyful, a
humane man is long-lived.”
子曰：知者樂水，仁者樂山。知者動，仁者靜。知者樂，仁者壽。
Lunyu 8/13 The Master said, “...When the world has the Way be visible but when it lacks the
Way be hidden. To be poor and lowly in a state with the Way is shameful; to be
rich and honored in a state without the Way is [also] shameful.”
子曰：...天下有道則見，無道則隱。邦有道，貧且賤焉，恥也；邦無道，富
且貴焉，恥也。
It is tempting to suppose that the clustering of template sayings within single Lunyu entries indicates that the Lunyu compilers thought of these sayings as templates. In other words, they were
540.The only other texts to include a high number of junzi/xiaoren sayings are the Liji (x20) and Huainanzi (x19),
both of which date to the Western Han. However, the relatively small size of the Lunyu means that it has a much
higher percentage of such sayings.
260
selected precisely in order to highlight certain frequently occurring tropes and to demonstrate
their rhetorical potential. That is not to say that the Lunyu compilers did not also value these sayings for their ideas. In particular, the first contrast at Lunyu 6/23 (“A wise man delights in water,
a humane man delights in mountains”) seems to have been a topic of great interest to authors in
the Han period.541 But in light of the Lunyu’s documented use as a textbook, we should also remain open to the possibility that these sayings reflect an interest in rhetorical instruction, in
teaching students the building blocks of intellectual discourse.
• Section #9: A Zengzi layer?
Since at least as far back as the seventeenth century, scholars in Japan, China, and the West have
agreed that the received Lunyu consists of multiple layers added at different moments in the
text’s history.542 However, as Oliver Weingarten has demonstrated, many markers which have
been used to distinguish these multiple layers do not hold up to scrutiny.543 These include the appearance of numbered lists, the use of zi 子 (The Master) versus Kongzi 孔子, and the use of certain grammatical particles (e.g., si 斯 versus ze 則). My own analysis of Lunyu parallels has likewise uncovered no evidence that smaller collections of Lunyu material might have circulated
independently in the pre-Han period. There is simply too little overlap among pre-Han Kongzi
sayings to discern any stable collection of Kongzi yue or zi yue material.
541.See Weingarten 2010, pp. 136–141, for a discussion of these parallels. For Weingarten, such passages are a
demonstration of the use of Kongzi and Kongzi disciple dialogues to dramatize textual exegesis.
542.For a summary of these efforts, see Weingarten 2010, pp. 29–34.
543.Weingarten 2010, pp. 34–57. Weingarten concludes (p. 56): “The relevance of most of the criteria that scholars
have so far used to distinguish textual layers in the Lunyu and to assign relative dates to them is debatable.”
261
Patterns of Lunyu parallels in Western Han texts, on the other hand, might point to the existence of material added to the Lunyu after its initial compilation. Recall Kaneto Mamoru’s observation (p. 185) that the “Kongzi shijia” and “Zhongni dizi liezhuan” include not a single saying
attributed to Zengzi despite the inclusion of fourteen Zengzi sayings in the received Lunyu.
Kaneto concludes from this (and other observations) that the Lunyu was compiled well into the
Western Han period and thus postdates the compilation of the Shiji. Without accepting Kaneto’s
larger conclusion, his observation alerts us to the possibility that the fourteen Lunyu Zengzi sayings were unavailable to Sima Qian because they were not in Sima Qian’s Lunyu.544
Certain peculiarities in the Lunyu Zengzi sayings themselves (listed Appendix 3:W) lend
some credence to this theory. Chief among these is the Lunyu’s representation (or lack thereof)
of the relationship between Kongzi and Zengzi.545 The Lunyu Kongzi interacts with his disciples
in various ways. Disciples ask questions, they voice opinions that Kongzi reacts to, and they
serve as fodder for Kongzi’s character judgments. However, the Lunyu Zengzi does not ask
Kongzi a single question in the text, nor does Kongzi comment on Zengzi or otherwise respond
to anything that Zengzi says or does.546 The single exception that proves the rule is Lunyu 4/15,
which I discussed in chapter one (p. 91) as an example of a Kongzi saying as object of commentary. In that passage, Kongzi addresses Zengzi with the words, “Can [=Zengzi]! My Way has a
single thread running through it” (參乎！吾道一以貫之。), and Zengzi subsequently explains
544.For this possibility, see also Watsuji 1948, pp. 335–336, and Tsuda 1946, pp. 67–68.
545.See also Haupt 2006, p. 81, on the possibility that Zengzi was not a disciple of Kongzi. She also observes that
Zengzi is not portrayed as a Kongzi disciple in the Mengzi and Xunzi (pp. 82–83). Mark Csikszentmihalyi (2004, p.
27, n. 33) also notes that in Hanshi waizhuan book eight (SBCK 8/15a–b) Kongzi criticizes Zengzi for excessive
filial piety.
546.On this point see also Van Norden 2002, p. 222. Lunyu 11/18 seems to comment on Zengzi but is not attributed
to Kongzi: “Chai [=Zigao] is stupid, Can [=Zengzi] is dull, Shi [=Zizhang] is excessive, You [i.e., Zilu] is coarse”
(柴也愚，參也魯，師也辟，由也喭。).
262
Kongzi’s meaning to other Kongzi disciples as “nothing but loyalty and empathy” (忠恕而已
矣). But note that the master-disciple relationship in their encounter is decidedly one-sided.
Whereas other Kongzi-disciple interactions in the Lunyu typically begin with the disciple providing the stimulus, here Kongzi simply addresses Zengzi and promptly leaves the stage, thereby allowing Zengzi’s interpretation to stand without comment.547 Lunyu 4/15 thus gives the impression that its author was more interested in establishing Zengzi’s superior understanding of
Kongzi than in involving Zengzi in a scene of instruction.
As it happens, the core Kongzi saying of Lunyu 4/15 also echoes a Kongzi saying at Lunyu
15/3:
The Master said, “Ci [i.e., Zigong], do you suppose that I know what I know by virtue of
having learned many things?” Zigong replied, “I do. Is that not the case?” The Master said,
“No. I tie [what I know] together with a single [thread].”
子曰：賜也，女以予為多學而識之者與？對曰：然，非與？曰：非也，予一以貫之。
Unlike Lunyu 4/15, Lunyu 15/3 does not elucidate Kongzi’s “single thread.” Also unlike Lunyu
4/15, Lunyu 15/3 appears in the “Kongzi shijia.”548 Given the dearth of Zengzi material in the
Shiji, this might suggest that Lunyu 4/15 was based on Lunyu 15/3 and was added to the Lunyu
by an editor keen on demonstrating Zengzi’s special grasp of Kongzi’s teachings.549
Oliver Weingarten has also argued independently in a study of Lunyu book one that “‘Zengzianist’ traditions” influenced the composition of that text.550 Weingarten locates a number of in-
547.Another peculiarity with Lunyu 4/15 is that it is the only complex dialogue in all of Lunyu book four, which
otherwise consists solely of simple zi yue sayings (save for Lunyu 4/26, the final saying in the book, which is a
Ziyou 子游 saying).
548.Shiji 47.1930.
549.Van Norden (2002) reaches a similar conclusion about the genesis of Lunyu 4/15.
550.Weingarten 2010, p. 69. Liu Zongyuan 柳宗元 (773–819) argued that Zengzi was the disciple primarily
responsible for creating the Lunyu on the evidence of Lunyu 8/3 and 8/4, which feature Zengzi on his deathbed. See
Liu Zongyuan ji 4.110–11.
263
triguing resonances between Lunyu book one sayings and Zengzi material in other texts, but his
strongest evidence is a parallel between Lunyu 1/1 and a Zengzi saying from Hanshi waizhuan
book two (with an additional parallel in Xunzi book 30):
<Lunyu 1/1>The Master said, “To practice what one has learned in timely fashion—is this not
a pleasure? To have a friend come from afar—is this not a joy? To not be resentful when
unrecognized by others—is this not noble?”
子曰：學而時習之，不亦說乎？有朋自遠方來，不亦樂乎？人不知而不慍，不亦君子
乎？551
<Hanshi waizhuan book 2>Zengzi said, “There are three sayings that a noble man can string
together and adorn himself with: the first is ‘do not be distant with those inside [the home] and
intimate with those outside; the second is ‘do not resent others for your own faults’; the third
is ‘[do not] invoke heaven after disaster befalls you.” Zigong said, “What are these?” Zengzi
said, “To be distant with those inside and intimate with those outside—is this not the opposite
[of what is proper]? To resent others for your own faults—is this not too far off the mark? To
invoke heaven after disaster befalls you—is this not too late?
曾子曰：君子有三言可貫而佩之：一曰：無內疏而外親，二曰：身不善而怨他人，三
曰：患至而後呼天。子貢曰：何也？曾子曰：內疏而外親，不亦反乎！身不善而怨他
人，不亦遠乎！患至而後呼天，不亦晚乎！552
The clearest parallel is the shared use of the thrice repeated pattern “is this not X” (bu yi X hu 不
以X乎). Also note both texts’ endorsement of not “being resentful” (yun 慍) or “resenting others” (yuan taren 怨他人) when faced with some difficulty. The Hanshi waizhuan’s use of guan
貫 (string together) in reference to a saying of unique importance also echoes the use of guan 貫
in Lunyu 4/15, where Zengzi explains the single thread running through Kongzi’s teachings. Assuming that the Hanshi waizhuan passage represents a pre-Lunyu tradition available to the Lunyu
compilers, it might have inspired two Zengzi-related selections: Zengzi’s “single thread” explanation of Kongzi’s Way, and the Zengzi-esque zi yue saying at Lunyu 1/1.
Why should a hypothetical pro-Zengzi and possibly post-Shiji compiler wish to elevate
551.The same pattern also appears in the Zengzi saying at Lunyu 8/7, although there it is only repeated twice and not
three times as it is here.
552.Hanshi waizhuan 2 (SBCK 2/6a).
264
Zengzi in this way? The frequent mention of the Xiaojing, a text attributed to Zengzi in the early
period,553 alongside the Lunyu in Han sources suggests one answer. Recall Emperor Zhao’s 昭帝
82 BCE edict (p. 171) that singled out the Lunyu and Xiaojing (together with the Shangshu) as
texts that he had learned but not yet mastered. In 74 BCE, Huo Guang communicated the future
Emperor Xuan’s 宣帝 qualifications for the throne by stating that he had “received instruction in
the Shi, Lunyu, and Xiaojing” (師受詩、論語、孝經) (p. 175). The Hanshu also records that
Liu Qu 劉去, appointed King of Guangchuan 廣川王 in 91 BCE, had mastered the Lunyu and
Xiaojing (together with the Yi), and that the future Emperor Yuan 元帝 had mastered the Lunyu
and Xiaojing by the age of twelve.554 The Kuang Heng 匡衡 memorial discussed in chapter two
(p. 175) likewise identified the Xiaojing together with the Lunyu as containing “the essentials of
the sages words and conduct” (聖人言行之要),555 and the “Yiwen zhi” listed the Xiaojing immediately after the Lunyu. If the Lunyu and Xiaojing were thought of as companion texts within the
context of the education of the Han royal family, our hypothetical pro-Zengzi Lunyu compiler
might have added Zengzi sayings to the text in order to enhance Zengzi’s standing and, ultimately, to encourage others to think of the Lunyu and Xiaojing as a unified curriculum. Read in conjunction with Lunyu 4/15, the Xiaojing might have offered insight into the “single thread” running through all of Kongzi’s teachings.
Conclusion
553.For the Xiaojing’s attribution to Zengzi, see Shiji 67.2305.
554.Hanshu 53.2428 & 71.3039.
555.The “Yiwen zhi” also mentions the Xiaojing alongside the Lunyu as one of the “ancient script-“ (guwen 古文)
texts recovered from the walls of Kongzi’s ancestral home. See Hanshu 30.1706.
265
These nine case studies have painted a complex picture of the Lunyu as a locus of multiple interests. The Lunyu compilers seem to have been especially interested in the assessment of character
and may have even conceived of the Lunyu as a kind of manual of character evaluation for an era
in which emperors claimed to personally oversee the selection of talented officials. The compilers’ handling of Zuozhuan sayings in particular reflected an interest in ren 仁 (humaneness) as an
abstract concept. They also included a number of sayings that resonated with other widely circulating proverb traditions (especially those in the Laozi), perhaps because they sought to establish
authoritative Kongzi versions of those earlier traditions. In this way, they asserted Kongzi’s ownership of the building blocks of early intellectual discourse at the same time as other Han scholars and statesmen were crediting Kongzi with the creation of the Five Classics. As Kongzi partisans, the Lunyu compilers scrubbed their Kongzi of any and all moral improprieties and also
used Kongzi’s own words to fix an authoritative portrait of the sage. Parallels with Chen and Cai
narratives show that they also engaged popular sub-genres of pre-Han Kongzi anecdotes; however, instead of participating in those traditions, they sought Kongzi sayings to stand as authoritative statements on those sub-genres. The inclusion of a relatively large number of widely attested
rhetorical tropes signaled the Lunyu compilers’ interest in the stock formulae of early rhetoric,
perhaps because they also conceived of the Lunyu as a kind rhetorical handbook. The observation that Lunyu Zengzi sayings do not appear in the Shiji led me to speculate that those sayings
were added at some point after the text’s initial compilation, perhaps in order to create an impression of the Lunyu and Xiaojing as companion texts. In short, the Lunyu compilers did what
other pre-Han and early Han authors probably did with Kongzi sayings: they treated them as a
flexible and dynamic medium of intellectual discourse. For them, Kongzi was an adaptable and
expansive reservoir of received wisdom.
266
This discussion has also revealed a similarly diverse array of compilation strategies. Some
Lunyu Kongzi sayings were extracted from longer dialogues, in a few cases with the vestiges of
those earlier contexts still intact. Other sayings with a legitimate pre-Han pedigree were developed into longer dialogues. The Lunyu compilers also attributed a number of anonymous sayings
to Kongzi and then personalized those sayings in a way that reinforced Kongzi’s ownership of
them. Some longer Kongzi dialogues were included as more or less complete episodes. Others
were mined for memorable lines without any consideration for the dialogue’s original intent,
while others were condensed and summarized in such a way as to capture the core message of
the source material. And I strongly suspect that a number of other Kongzi sayings were composed originally for the Lunyu, although identifying such material is impossible.
What can this analysis of Lunyu parallels tell us about the circumstances of the Lunyu’s compilation? One way to approach this question is to ask who in the early Western Han might have
been capable of putting together a compilation like the Lunyu. Although pre-Han Lunyu parallels
are few and far between, there are enough of these parallels in the Zuozhuan, Mengzi, “Zi yi,”
Laozi, and elsewhere to suggest that the Lunyu compilers had direct access to a number of
sources. Judging from its appearance in multiple manuscript finds and its frequent citation in,
e.g., the Huainanzi, the Laozi in particular seems to have circulated widely in the Western Han
period (at least relative to other texts). While quotations of the received Mengzi are relatively
rare in texts from the period, Sima Qian wrote of Mengzi and several other masters that “many
people nowadays have their writings” (世多有其書)—perhaps the Lunyu compilers also had access to a Mengzi or proto-Mengzi.556 The Zuozhuan, on the other hand, is an exceedingly large
556.Whether or not Mengzi’s “writings” (shu 書) refers to something resembling the received Mengzi is another
question entirely.
267
compilation which presumably did not have a large circulation, although smaller collections of
Zuozhuan material may have had a larger audience.557
Earlier I suggested that the clustering Zuozhuan parallels in Lunyu 12/1 and 12/2 indicate a
close engagement with a Zuozhuan text. Who in the Western Han might have owned a copy of
the Zuozhuan, Mengzi, Laozi, etc.? The list of potential Lunyu sponsors begins with the Han emperors, who possessed multiple libraries at Chang’an,558 followed by regional kings like Liu An
劉安 (179–122), the King of Huainan 淮南王, and Liu De 劉德, King of Hejian 河間王 (r. 155–
130/129),559 both of whom reportedly attracted a great number of scholars to their courts,
amassed their own book collections, and sponsored their own compilations.560 Of these two figures, Liu De is by far the more intriguing candidate, at least judging from his Hanshu biography
(Appendix 3:X), the most intriguing detail of which is the short list of texts he reportedbly possessed: the Zhouguan 周官 (Zhou Offices), Shangshu 尚書 (Exalted Documents), li 禮 (ritual)
texts, Liji 禮記 (Ritual Records, or possibly generic “ritual records”), Mengzi, Laozi, Mao Shi 毛
詩 (Mao Odes), and Zuo Chunqiu 左春秋 (Zuo’s Spring and Autumn Annals, i.e., the Zuozhuan).
Of these, the Liji (again assuming that Liu De’s li and liji texts are related to the received Liji),
557.For one example of such a text, see the “Chunqiu shiyu 春秋事語” manuscript from the Fuyang find. See Han
Ziqiang 2004, pp. 165–205.
558.Csikszentmihalyi 2002, p. 148: “This was also the period [Emperor Wu’s reign] that saw the construction of
imperial book repositories and attempts to take a census of all book titles....The role of Confucian texts in the
imperial examination system and the establishment of official positions devoted to their study was probably the most
important single catalyst for consolidating and editing the texts associated with Confucius.” For the central book
collections, see the Liu Xin fragment preserved at Hanshu 30.1702, n. 10: “Liu Xin’s Seven Summaries said,
‘Outside [of the palace] there are the holdings of the Grand Master of Ceremonies, the Grand Scribe, and the
Erudites, and within [the palace] there are the archives of the Extended Pavilion, the Expansive Chamber, and the
Secret Room” (劉歆七略曰：外則有太常、太史、博士之藏，內則有延閣、廣內、祕室之府。).
559.For this suggestion, see also Kaneto 1974, p. 16. Takeuchi Yoshio (1937) argued that the earliest stratum of the
Lunyu (books 2-9) was first compiled at the court of Liu De. However, Takeuchi also maintained that this stratum
represented Kongzi’s original teachings as transmitted by Zengzi via Mengzi.
560.For a discussion of the texts associated with Liu An’s court, see Le Blanc 1985, 41–52.
268
Laozi, Zuozhuan, and Mengzi exhibit a number of obvious Lunyu parallels.561 It is tempting to
speculate that patterns of Lunyu intertextuality reflect the immediate textual environment of Lunyu compilers working under the auspices of an emperor or king like Liu De. I indulge that temptation in an epilogue.
561.While the Mao commentary to the Shi has no Lunyu parallels among its five Kongzi sayings, there are a handful
of parallels that might suggest some sort interaction between the two texts. See the entries for Lunyu 5/21, 13/28,
15/27, and 16/2 in Appendix 2:X.
269
CONCLUSION:
The Place of Confucius and the Analects in Early Chinese Thought
In this dissertation I have posed a series of challenges both to the traditional view of the Lunyu
and to the standard approach to the sources of Kongzi’s life and thought. Chapter one’s survey
revealed the limitations of viewing early Kongzi sayings through the narrow lens of the Lunyu, a
text whose preference for independent Kongzi sayings appears idiosyncratic when considered
against the corpus of Kongzi sayings at large, and particularly those in pre-Han and early Western Han sources. I also developed the argument that Kongzi sayings wherever they appear need
not be read as sources for the historical Kongzi, or even as material transmitted by generations of
devoted disciples in a Confucian “school” (jia 家) or disciple lineage. Kongzi yue marked a
medium that was useful and meaningful on its own terms, and in particular as a source of comments and commentaries on pre-existing textual traditions.
Chapter two’s chronological survey of Kongzi quotation practice showed that early authors
did not begin to rely on the Lunyu as an authoritative source of Kongzi sayings until well into the
Western Han period. Prior to that point, early authors did not reference a Lunyu text or any other
bounded collection of Kongzi material, nor did they privilege Lunyu Kongzi sayings over nonLunyu Kongzi sayings. From the distribution of Lunyu Kongzi sayings in early Western Han versus mid- to late Western Han texts, and especially in Han edicts and memorials, I pinned the beginning of this transition towards the end of Emperor Jing’s reign and the beginning of Emperor
Wu’s reign, perhaps between 150 and 130 BCE. Although close parallels between the Lunyu and
pre-Han Kongzi sayings have been cited as evidence that something resembling the received
Lunyu circulated in the period, in fact such parallels are few and far between and are far outnum270
bered by anonymous or differently attributed parallels.
In chapter three I set out to demonstrate the plausibility of reading the Lunyu as a product of
the Western Han period. In a series of nine case studies I read Lunyu Kongzi sayings against
their extant parallels to as selections from earlier material. I then used the selection strategies
which emerged from this analysis to suggest a few conclusions about the interests and motivations of the Lunyu compilers. I ended with the suggestion that the Lunyu might have been sponsored by King Xian of Hejian, a figure with the requisite interests and resources to produce such
a text.
These conclusions do not begin to exhaust the implications of understanding the Lunyu as a
product of the Western Han period. The traditional view of the Lunyu does a lot of work in
standard accounts of early Chinese thought: by asserting the Lunyu’s ancient provenance, it anchors the standard narratives of the development of early Chinese thought and promises direct
access to Kongzi as the earliest and most influential thinker in the Chinese tradition. As Mark
Csikszentmihalyi has written, “[m]inimizing the authorship role of Kongzi has the effect of
pulling the rug out from under the usual narrative of what is often called the ‘history of thought’
(sixiang shi 思想史) of early China.”562 What features of that narrative must we reconsider, and
how should the narrative be rewritten in order to accommodate both the revised dating of the
Lunyu and the picture of Kongzi quotation practice presented here?
1) There is no solid evidence for the existence of a Lunyu text prior to the Western Han period.
562.Csikszentmihalyi 2004, p. 25.
271
First, a caveat: nothing in this dissertation completely precludes the possibility that a text resembling the received Lunyu circulated prior to the Western Han period. The analysis of quotation
patterns can only reveal what the authors of extant texts chose to attribute to Kongzi; it cannot
reveal whether a Lunyu text actually existed in the Warring States period. Perhaps the text was
transmitted esoterically, or perhaps Wang Chong was correct in thinking that the text had been
lost at some point in the Warring States period only to be rediscovered in the Western Han. Nor
can I entirely rule out the possibility that future manuscript finds will turn up a legitimately preHan proto-Lunyu manuscript, e.g., a collection of Kongzi sayings with roughly the same content
and format as the sayings of the received Lunyu.
But I would not bet on it. While the arguments of this dissertation do not completely disprove
the standard dating of the Lunyu, they do reveal its shortcomings. One way to approach this
problem is to ask what kinds of evidence would establish the existence of a pre-Han Lunyu (aside
from the discovery of a pre-Han Lunyu manuscript). Such evidence might include:
• explicit mentions of a “Lunyu” text plus information connecting that “Lunyu” to our Lunyu;
• a consistently higher ratio of Kongzi sayings with Lunyu parallels versus Kongzi sayings
without Lunyu parallels;
• the clustering of Kongzi sayings with Lunyu parallels around specific Lunyu books, which
would indicate that those clusters circulated independently prior to being included in the
received Lunyu;
• a degree of variability between the Kongzi sayings of a pre-Han Lunyu and those of the received Lunyu which does not exceed the variability observed between other received
texts and their pre-Han counterparts (e.g., the various Laozi and “Zi yi” manuscripts);
• the observation that only Lunyu Kongzi sayings were treated as relatively authoritative
272
sources of Kongzi’s life and thought, e.g., as objects of commentary or as proof texts.563
In my reading of the relevant sources I have found no such traces of the Lunyu’s existence prior
to ca. 150 BCE. To continue to insist on the validity of the traditional view based on the still-notentirely-disproven possibility of the existence of a pre-Han Lunyu is to run the risk of building
our “knowledge” of early China on the flimsiest of foundations.
But let us suppose that archaeologists were to discover a legitimately pre-Han Lunyu manuscript. Given the remarkable dearth of Lunyu Kongzi sayings in the pre-Han corpus, such a find
would still not be the supremely authoritative and uniquely authentic source posited by the Han
bibliographers, because pre-Han authors by and large did not exhibit an awareness of any stable
source of Kongzi sayings. Whatever else, the story of the Lunyu as an authoritative source of
Kongzi sayings—in other words, the story of our Lunyu—begins in the Western Han.
2) The Lunyu is, at best, a problematic source of the pre-Han period.
The conclusion of the foregoing section comes with another caveat: just because the Lunyu did
not circulate as a text in the pre-Han era does not mean that all Lunyu material originated in the
Western Han.564 As established in chapters two and three, some Lunyu sayings do have a legiti-
563.Here I do not address what might be called “the argument from philosophy,” the notion that the Lunyu can be
dated to the pre-Han era because its ideas reflect a pre-Han as opposed to a Han context. For that argument, which I
hope to address in a future publication, see Goldin 2011 and Slingerland 2000. In brief, my primary objection to
such an argument is methodological. Since our sense of what constitutes an earlier as opposed to a later idea has
been shaped by traditional chronology (e.g., the Lunyu preceded the Mengzi which preceded the Xunzi), defending
traditional chronology based on the dating of ideas seems an unavoidably circular enterprise.
564.On this point see also Tsuda 1946, pp. 14–15. Also compare Makeham 1996, p. 14: “...Yet this is by no means
to imply that the contents of these twenty-one pian were a Western Han fabrication. There is no reason to suspect
that much if not most of the material contained in these twenty-one pian did not ultimately derive from the early
Warring States period.”
273
mate pre-Han pedigree as Kongzi sayings, and many more Lunyu sayings have a history as sayings attributed to other figures or to no one at all. But if the Lunyu was compiled in the Western
Han, inclusion in the Lunyu in and of itself cannot be taken as a mark of a saying’s pre-Han origins. The only reliable way to determine whether a given Lunyu entry originated in a pre-Han
context is to identify parallels in texts reliably dated to the pre-Han period. Even when such parallels can be found, there is still the additional question of whether the Lunyu preserves the earlier version more or less unchanged, or whether the Lunyu compilers elided precisely those features which made it interesting as an artifact of the pre-Han context.
Consider the example of Lunyu 2/17 discussed in chapter three (p. 225), whose parallels in
the Xunzi and Hanshi waizhuan are longer dialogues between Kongzi and Zilu in which the saying at Lunyu 2/17 (“take knowing as knowing and not knowing as not knowing—this is knowing”) is but one unemphasized element in Kongzi’s instruction. Assuming that Xunzi book 29 is
truly a product of the third century BCE, the idea and wording of Lunyu 2/17 would seem to date
to the pre-Han period. But the decision to decontextualize the earlier dialogue and treat Lunyu
2/17 as an independent Kongzi saying dates to the Western Han. The content is pre-Han, its presentation and emphasis is Han.
How we choose to interpret Lunyu 2/17 will depend first and foremost on the questions we
ask of the text. A linguist looking to build a corpus of pre-Han language would be justified in including Lunyu 2/17 in her data set, and a historian of ideas could read it as evidence that the notion of privileging second-order knowledge was current in the pre-Han era (again, assuming that
the Xunzi parallel dates to the third century). But given the Lunyu compilers’ decision to extract
only a single line and discard everything else, thereby obliterating the very ideas that motivated
the anecdote, my preference would be to read Lunyu 2/17 as a “Western Han text”* with an
274
asterisk.
3) There is no solution to the “Kongzi problem,” i.e., no way to recover the life and thought of
the historical Kongzi from extant Kongzi material.
Let us step out of the early Chinese context for a moment to consider the parallel case of
Socrates (ca. 469–399 BCE).565 From an early China scholar’s perspective, the sources of
Socrates’s life and thought are an embarrassment of riches. Socrates’s existence is confirmed by
one contemporary fifth-century source, Aristophanes’ (ca. 446–386 BCE) Clouds, as well as a
number of Socratic dialogues written in the decades immediately following Socrates’s death in
399 BCE. Although Plato’s (420s–348/47 BCE) dialogues are the most well known of these, other associates of Socrates also participated in the genre, including Aeschines of Sphettus (430/20–
after 375/6 BCE; seven dialogues, all lost), Phaedo of Elis (b. 418/16 BCE; two dialogues, both
lost), Euclides of Megara (450/35–c. 365; six dialogues, all lost), Antisthenes (c. 445–c. 365; a
number of dialogues, all lost), and Xenophon (430–354), whose Symposium, Memorabilia, Oeconomicus, and Apology are extant.566 Aristotle names yet another figure, a certain Alexamenos
of Teos, as the first person to have penned a Socratic dialogue.567 Excavators of the Athenian
agora in the 1950s even claimed to have confirmed a detail from Xenophon’s account of
565.Christiane Haupt (2006, pp. 18–20) has argued along similar lines from a comparison between Kongzi and
Jesus. For a summary of the problems involved in reconstructing the historical Jesus, see Tuckett 2001; for a
summary of past reconstruction efforts, see Paget 2001; for a discussion of the quest for the “real” Jesus, see Watson
2001. Especially useful is Tuckett’s discussion (pp. 132–137) of the criteria various scholars have derived for
identifying historically accurate sources. McClymond’s (2001) discussion of the sources of Jesus’s life and thought
is also relevant in the Chinese context.
566.Döring 2011.
567.For this fragment, see Döring p. 25, n. 2.
275
Socrates in the Memorabilia and from Diogenes Laertios’s Lives of the Philosophers after they
uncovered what appeared to be a leather-working shop together with a cup engraved with the
name “Simon.” They speculated that this Simon was the same Simon said to have owned a
leather shop frequented by Socrates and who supposedly made notes of his conversations.568
Extant sources for Socrates more or less agree on a few basic biographical details: Socrates
was an Athenian, a thinker, a teacher, and a conversationalist. But scholars of the period continue
to disagree about the historical value of these often contradictory accounts, particularly with respect to Socrates’s doxography. The earliest source, Aristophanes’s Clouds, is an obvious parody,569 and many Platonic dialogues give the impression that Plato was less interested in describing Socrates’s philosophy than in using the figure of Socrates the dialectician as a useful prop.
Occasionally, Plato even drops hints that his version of events might be less than completely accurate. In the Phaedo, the dialogue which purports to recount the circumstances of Socrates’s
death, he even has the narrator go out of his way to note that Plato himself was absent due to
illness.570
The proliferation of Socratic dialogues in the fourth century is one hint that these texts were,
first and foremost, a dynamic genre of intellectual discourse; they were not intended to be read as
historically accurate accounts. As some scholars of the period have argued,
[G]iven that Plato, like Xenophon and the other Socratics, were writing in a literary genre well
described as “biographical experiments” that aim at “capturing the potentialities rather than
568.See Döring 2011, pp. 34–36, and Lang 1978, p. 16. For the primary sources, see Xenophon (Memorabilia 4.2)
and Diogenes Laertios (Lives of the Philosophers, 2.13.122). Simon was also the name of a dialogue written by
Phaedo of Elis.
569.Dorion 2011, p. 6 (Cambridge Companion to Socrates): “It is probably impossible to reconstruct the ideas of
the historical Socrates from Aristophanes’ The Clouds, not only because the very genre of comedy lends itself to
exaggeration and even excess, but also because there is good reason to believe that Socrates’s character in The
Clouds is really a composite figure whose traits were gathered not only from Socrates himself but also from the
physiologoi and the sophists.”
570.Phaedo 59b.
276
the realities of individual lives” (Momigliano 1993: 46), what hope is there for reconstructing
the historical Socrates from these representations? The representations conflict at the most
basic level: Socrates affirms and denies that the good is pleasure (Plato, Gorgias 495a–99b,
but cf. Protagoras 351b–e, 354de); Socrates does and doesn’t investigate questions of natural
science (Aristophanes, Clouds 217–33; Aristotle, Metaphysics A.6.987b1–3; Xenophon,
Memorabilia 1.1.11–16, 4.7.2–10; Plato, Phaedo 96d–99e, but cf. Apology 26de); Socrates
disavows and avows having knowledge (Plato, Apology 21b–23b, Theaetetus 150cd, but cf.
Apology 29b). So why suppose that the Socrates of Plato’s early dialogues was the historical
Socrates, rather than the Socrates of Xenophon’s Socratic writings, or the Socrates of
Aeschines, or Aristippus, or indeed of the hostile witness Aristophanes?571
These Western classicists’ willingness to acknowledge the impossibility of recovering the
thought of Socrates, despite Socrates’s exalted status in the Western tradition and the (relative)
wealth of nearly contemporaneous sources at their disposal, is exemplary. In the absence of reliably dated, roughly contemporaneous records of Kongzi’s life and thought, students of early China have even less hope of sorting the historically authentic Kongzi from later “biographical experiments,” or even of determining whether extant texts preserve any accurate information about
Kongzi. Like Socratic dialogues, Kongzi sayings were a useful medium regardless of their
provenance or historical authenticity, and we need not suppose that early audiences would have
understood Kongzi yue or zi yue in any text as the mark of a historical record.
At the same time, Socrates probably would not have inspired later writers had he not possessed an extraordinary charisma, or at least an extraordinary reputation. Applying the same logic to Kongzi, we might infer that someone triggered the tremendous interest in a figure known as
“Kongzi” or “Zhongni” or “Kong Qiu.” But the story of how this figure became a household
name in ancient China remains a mystery.
4) Kongzi quotation practice did not necessarily depend on a Kongzi intellectual lineage or dis-
571.Ahbel-Rappe and Kamtekar 2006, pp. xiv-xv.
277
ciple transmission.
Whatever the relationship between the historical Kongzi (assuming there was such a person) and
the words and deeds attributed to him in later texts, by the fourth and third centuries BCE various
kinds of Kongzi material were in circulation even outside of Kongzi’s supposed home in the
state of Lu.572 From the scattered “Kongzi learned of this and said” (Kongzi wen zhi yue 孔子聞
之曰) comments in the Zuozhuan and Lüshi chunqiu to the commentaries of the “Kongzi shilun”
and Zhouyi, and from the Kongzi yue proof texts of the Mengzi to the various Kongzi dialogues
in the Shanghai Museum manuscript corpus and elsewhere, Kongzi sayings were being put to diverse uses by numerous pre-Han authors. Within the Shanghai Museum corpus alone, a collection which has been tentatively dated to ca. 300 BCE, we find self-contained Kongzi anecdotes
and dialogues and even a list of quotations attributed to an anonymous “master” (zi 子) who may
or may not have been Kongzi. If, as seems likely, similarities with the Guodian find indicate that
the Shanghai Museum manuscript corpus originated in the south, by the late fourth century BCE
Kongzi material had spread hundreds of miles southward from his home in the state of Lu (modern-day Shandong province) to the Chu region (modern-day Hubei).
Even granting that the Lunyu did not exist until the Western Han period, the traditional view
suggests a mechanism by which Kongzi sayings and stories became so prevalent: they were circulated and transmitted by people who self-identified as disciples in a Kongzi or Kongzi disciple
lineage. These transmitters could have been proselytizers of a sort, people like Mengzi who sang
572.Zhu Weizheng (2007, pp. 169–170) sketches a four-stage development in which Kongzi was elevated from a
mere xianren 賢人 (man of accomplishment) to a shengren 聖人 (sage), from shengren to suwang 素王 (uncrowned
king), and from suwang to jiaozhu 教主 (religious founder).
278
Kongzi’s praises and credited him with having found the Way (dao 道) to cure society’s ills. According to that theory, it was these disciples’ direct connection to Kongzi and their supposed fidelity to preserving Kongzi’s memory which guaranteed the authenticity of Lunyu Kongzi
sayings.573
But what evidence does early Kongzi material really provide for the devoted disciple model
of textual transmission? Kongzi sayings and stories routinely represent Kongzi in conversation
with a recurring cast of disciples. Yan Hui and Zilu and Zigong appear as Kongzi disciples in so
many early texts, one might argue, because Kongzi actually taught these men, who then passed
on his teachings to subsequent generations while preserving the original instructional contexts.
Some Kongzi quoters do seem to have expressed an interest in safeguarding Kongzi’s memory.
Recall Zigong’s 子貢 criticism of the eulogy delivered by Duke Ai of Lu 魯哀公 upon Kongzi’s
death in the Zuozhuan (p. 99), Mengzi’s effort to defend Kongzi’s reputation and exclude spurious (according to him) Kongzi sayings at Mengzi 5A/4 and 5A/8 (p. 117), and Youzi’s criticism
of Zengzi in “Tan gong” (p. 46). There are also instances in which a Kongzi quoter indicates that
he has “heard” (wen 聞) a Kongzi saying from Kongzi or a Kongzi disciple, as when a certain
Yuezheng Zichun 樂正子春 in the Lüshi chunqiu says that he “has heard it from Zengzi, who
heard it from Zhongni” (聞之曾子，曾子聞之仲尼).574 To these examples we could add the various episodes in which Kongzi enjoins his disciples to ji 記 (“take note of” or perhaps even “note
573.In his study of early masters (zi 子) texts, Mark Edward Lewis (1999) cited the Lunyu as his primary example of
the master–student–text triad, whereby masters like Kongzi was constituted by groups of disciples who possessed
special written records of their master’s teachings, students asserted their identity through their devotion to a master
and their possession of special records, and texts representing masters were developed and maintained by devoted
disciples. While Lewis’s model might very well apply to other early masters, we need not appeal to such a model to
explain the genesis of the Lunyu or of other kinds of Kongzi material.
574.Lüshu chunqiu 14/1, “Xiao xing 孝行” (SBCK 14/3a). However, it seems to me that such sayings are rather rare.
In the Lunyu, only 17/7, 19/17, and 19/18 feature disciples repeating words they had heard from the Master.
279
down”) his words (p. 51).
On the other hand, the seeming universality of Kongzi quotation practice in the pre-Han period coupled with the striking lack of agreement among pre-Han Kongzi quoters about what
Kongzi actually said hint that Kongzi quotation practice was not restricted to a small group of
self-identified Kongzi disciples. The geographical scope of Kongzi sayings is also suggestive.
Kongzi dialogues in several pre-Han texts also have him traveling to and from other Spring and
Autumn-era polities, especially Qi 齊, Wei 衛, and Chu 楚 and the infamous region “between
Chen and Cai.” According to the Lüshi chunqiu,
Kongzi drifted all around within the [four] seas seeking out one contemporary ruler after
another. From Qi to Wei he saw more than eighty different lords.
孔子周流海內，再干世主，如齊至衛，所見八十餘君。575
Numerous early texts also portray Kongzi as having “heard of” (wen 聞) and commented on
events in nearly every Spring and Autumn-era polity. Moreover, the use of Kongzi quotations
even by a staunchly anti-Kongzi figure like Mozi indicates that Kongzi quotation practice
transcended ideological boundaries, and that Kongzi sayings were a shared intellectual currency.576 Sima Qian suggests as much in the “Kongzi shijia” when he writes, “The words Kongzi uttered while serving as an official or hearing cases could be shared with others—there was nothing exceptional about them” (孔子在位聽訟，文辭有可與人共者，弗獨有也。).
The vast majority of authors who attributed sayings to Kongzi do not seem to have worried
575.Lüshi chunqiu 14/7, “Yu he 遇合” (SBCK 14/18b; KZJY 12.28/p. 262). In contrast, Nylan & Wilson (2010, p.
30) argue that Kongzi was originally a local culture hero from Lu. While Kongzi’s close association with the state of
Lu 魯 in pre-Han anecdotal literature is undeniable, quotations like this one complicate that conclusion.
576.It is for this reason that I cannot entirely agree with Nylan and Wilson (2010, pp. 29–30) that, prior to the Qin
unification, “it was never clear that Kongzi would eventually be hailed as unique and the most important figure in
Chinese history.” Even in the pre-Qin period the prevalence of Kongzi sayings and Kongzi’s role as commenter/
commentator extraordinaire set him apart from other intellectual authorities. The Han apotheosis of Kongzi, and
especially his identification as the author/editor/compiler of the Classics, only makes sense in light of this context.
280
about their authenticity. The “man of Chu lost his bow” anecdote from the Lüshi chunqiu (p. 57)
and the Kongzi dialogues from the Zhuangzi, all of which were labeled “imputed words” (yu yan
寓言) by Sun Xingyan in the early 19th century, are unique only insofar as they wear their artificiality on their sleeve. Even the author of Mengzi 5A/4 only criticized a Kongzi saying on the
grounds that it was unworthy of Kongzi’s reputation; nowhere did he claim that some Kongzi
sayings were more historically accurate than others. The Mengzi author’s primary concern was
upholding Kongzi’s image for his contemporaries, a project which did not depend on recovering
the historical Kongzi or transmitting Kongzi sayings verbatim. One of the few passages in early
Chinese texts to suggest an interest in “genuine” (zhen 真) Kongzi sayings appears in Xunzi book
six, “Fei shier zi 非十二子” (“Against Twelve Masters”): “[Some men] ornament their words in
order to gain others’ respect, saying, ‘These are the genuine sayings of the former junzi. Zisi
chanted them, and Mengzi [sang] in harmony with them’” (案飾其辭，而祇敬之，曰：此真先
君子之言也。子思唱之，孟軻和之).577 However, the author of this passage intended it as a
criticism and apparently counted on his audience finding such pious claims preposterous. As Dorion has noted with respect to early Socrates material, “One of the reasons the Ancients never debated the Socratic problem is because they fully recognized the fictional nature of the logoi
sokratikoi.”578 I see little reason to suppose that ancient purveyors and consumers of Kongzi material would have felt any differently.
To be clear, my argument is not that Kongzi sayings were never transmitted word-for-word
577.Xunzi 6 (SBCK 5/14b).
578.Dorion 2011, p. 13. Dorion goes on to list a handful of examples in which Greek and Roman authors
acknowledge the fiction of Socratic dialogues. Ahbel-Rappe and Kamtekar (2006, p. xv) credit Paul Vander Waerdt
(1994, p. 3) with “having voiced his skepticism about the sources’ ability to tell us much about the historical
Socrates [and] suggest[ing] that these sources might be better used as guides to the thinking of their authors or for
the recovery of philosophically brilliant portraits of Socrates.”
281
by self-identified Kongzi disciples in the pre-Han period. My proposal is that students of early
China should reconsider the default approach to the interpretation of early Kongzi material. Especially when dealing with pre-Han Kongzi material, the faithful transmission of Kongzi sayings
must be demonstrated, not assumed. And if Kongzi quotation practice did not depend on masterdisciple transmission, there may also be a need to reconsider the centrality of the master-disciple
model of textual transmission in early texts generally.
5) The Lunyu is an imperial text.
By “imperial text” I do not just mean that the Lunyu was compiled in the imperial period. I
would also like to suggest that the success of the Lunyu project, understood as the effort to create
a delimited set of authoritative Kongzi sayings, hinged on political centralization. It is probably
not a coincidence that the Lunyu is first mentioned in the historical records of Emperor Wu’s
reign, a time when the Han imperium was consolidating and expanding its power to an unprecedented degree. Kongzi quotation practice prior to the advent of the Lunyu was so decentralized
and unruly that no source of Kongzi sayings could have become authoritative without the backing of the highest political and/or intellectual authorities. The fact that the earliest recorded mentions of the Lunyu in the Han histories are in the context of the education the Han royal family
likewise points to the Han imperium as the force responsible for the text’s initial success. Eventually, late Western Han Lunyu transmitters and other scholar-officials like Kuang Heng, Liu Xiang, and Yang Xiong saw to it that the Lunyu’s authority extended beyond the royal family to the
Han elite generally, in part by advancing the argument that the Lunyu provided unfettered access
to Kongzi as the sage-author of the Classics. But these scholars’ contributions were restrospec282
tive rationalizations; in my view, the Liu 劉 clan deserves the lion’s share of the credit for the
Lunyu’s status in the Chinese tradition.
One of the enduring biases in the study of early Chinese thought is the privileging of the
Warring States era over the early imperial period.579 This preference can be explained in part by a
natural preference for the most ancient sources of ancient Chinese thought, the assumption being
that these sources are somehow more foundational than texts written in later periods. Thus, the
teachings of, say, the historical Kongzi or Mozi are often seen as more profound or essential than
the writings of later Confucians and Mohists. A perceived correlation between political decentralization and intellectual freedom also contributes to an anti-imperial bias. The diversity and vibrancy of Warring States-era thought as evidenced in the texts attributed to the eponymous Warring States masters has been contrasted with the situation under the Qin and Han empires when
(so the story goes) the imperial orthodoxy stifled intellectual innovation.580 (This view of early
Chinese intellectual history seems to have drawn inspiration from the history of Greco-Roman
thought, in which democracy has been credited with the intellectual vibrancy of fifth-century
Athens and the Roman Republic, and Hellenistic monarchy and the Roman Empire with the end
of the Golden Age of Greek and Roman thought.) The traditional view of the Lunyu has abetted
this bias by offering the Lunyu as our best source of the figure who ushered in a Golden Age of
579.For useful correctives see Csikszentmihalyi 2006, pp. xx–xxiii, and Michael Nylan’s (1999) argument against
the “Han orthodox synthesis” and “victory of Confucianism” model.
580.For a recent statement, see Pines 2009, p. 3. A classic formulation of this thesis belongs to Karl Jaspers (1953,
p. 4): “Corresponding to this new spiritual world [of the Axial Age], we find a sociological situation showing
analogies in all three regions [i.e., China, India, and the Mediterranean world]. There were a multitude of small
States and cities, a struggle of all against all, which to begin with nevertheless permitted an astonishing prosperity,
an unfolding of vigour and wealth. In China the small States and cities had achieved sovereign life under the
powerless imperial rulers of the Chou dynasty; the political process consisted of the enlargement of small units
through the subjection of other small units. In Hellas and the Near East small territorial units—even, to some extent,
those subjected by Persia—enjoyed an independent existence. In India there were many States and free cities.”
283
Chinese philosophy, and whose biography stands as a testament to the political fragmentation of
the age.
From this perspective, labeling the Lunyu an imperial text is not simply a matter of removing
one of the most cherished sources of Warring States thought. It is also a direct challenge to reconstructions of early Chinese thought which privilege the pre-imperial over the imperial without
subjecting the chronology of ostensibly pre-imperial texts to critical scrutiny. If the Lunyu
Kongzi is a Han creation, albeit one which resonates in certain ways with pre-Han Kongzi traditions, there is clearly a need to reconsider the dating of other received texts from the Warring
States era in light of the possibility that they owe much of their form and content to early imperial editors and compilers. How the history of Warring States thought should be rewritten to accommodate a Han perspective is a question beyond the scope of this dissertation. Given the ideological diversity of pre-Qin manuscript finds, one possibility is that histories of Warring States
thought will come to see the syncretism or eclecticism evident in texts like the Lüshi chunqiu and
Huainanzi not just as hallmarks of early imperial thought, but as pervasive features of early Chinese thought generally. Instead of presenting the story of Warring States thought as a parade of
masters and “-isms,” scholars may also come to see the genre of the Warring States master text
as part of a Han effort to organize received textual traditions around figures who, like Kongzi,
possessed a reputation and a symbolic status but not necessarily a well defined corpus of teachings. In short, just as the traditional view of the Lunyu has encouraged generations of scholars to
view Kongzi as the prototypical master (zi 子) and the Lunyu as the prototypical master text, the
Kongzi who emerges from Kongzi quotation practice together with the theory of a Western Han
Lunyu might serve as a model for the study of pre-Han masters and master texts generally.
284
AN IMAGINATIVE EPILOGUE:
The Compilation and Rise of the Analects, circa 150 BCE–82 BCE
In chapter two I presented a number of arguments for revising the traditional dating of the Lunyu,
and I concluded by suggesting the early part of Emperor Wu’s reign as an approximate date of
compilation. Unfortunately, the available evidence does not allow us to say with any certainty
when exactly the Lunyu was composed, who composed it, or under what circumstances. Barring
some spectacular manuscript find, we will never know everything we would like to know about
the circumstances of its creation, nor will we possess more than circumstantial evidence for our
theories of the text’s genesis.
Be that as it may, ancient Chinese texts did not exist in a vacuum, and our theories of textual
formation have any number of implications for what real-world people were actually doing with
their texts. Testing those implications against our limited knowledge of the relevant social, intellectual, political, and material contexts is one of the only ways we have to test the plausibility of
our theories and, in the process, to suggest avenues for future research. To that end, in this section I develop one possible scenario for the Lunyu’s genesis in the Western Han. My account is,
in part, an effort to reverse-engineer the circumstances of its creation: given what we know about
its possible sources and the editorial decisions that shaped the text as we have it today, what can
we surmise about the Lunyu compilers and their immediate context? And given what we know
about textual production in the early Western Han, who might have initiated such a project, and
for what purpose(s)? Although necessarily hypothetical, the story I will tell hews as much as possible to the “facts” of Western Han history as attested in the early histories, imperfect sources
though they may be.
285
*
*
*
One setting in which we could imagine a Lunyu text being compiled is the court of Liu De 劉德,
the figure posthumously known as King Xian of Hejian 河間獻王 (r. 155–130/129 BCE). Not
long after being enthroned as king by his father, Emperor Jing 景帝 (r. 156–141),581 in 155 BCE,
Liu De 劉德 began to establish a reputation as a patron of traditional learning and an avid collector of texts. The rewards he offered for donated texts, all of which he had faithfully transcribed
so that donors could keep copies for themselves, guaranteed that Liu De’s collection soon rivaled
that of the imperial court.582 Liu De was an equally avid collector of intellectuals, who staffed his
bureaucracy, maintained his library, and created a vibrant intellectual atmosphere at his court.583
Liu De apparently relished one of the prerogatives of intellectual patrons in the period: overseeing the creation of new texts:
During Emperor Wu’s reign, King Xian of Hejian loved the Ru, and together with Master Mao
and others selected [passages] from the Zhou Offices and the masters’ discussions of musical
matters to compose the Record of Music. He also presented the “eight-row dance,” which was
not far off from Mr. Zhi’s [version].
武帝時，河間獻王好儒，與毛生等共采周官及諸子言樂事者，以作樂記，獻八佾之舞，
與制氏不相遠。584
There is little evidence to suppose that Liu De’s Yue ji 樂記 (Record of Music) was the same
“Yue ji” included in the received Liji.585 Regardless, the description of this text as having been
581.Many of these details are drawn from his Hanshu biography in book 53, “Biographies of the Thirteen Kings of
[Emperor] Jing” (景十三王傳). See Hanshu 53.2410–2411; for a complete translation, see Appendix 3:X.
582.Hanshu 53.2410: Liu De’s collection consisted of “ancient-text pre-Qin old writings..., all of them canons and
commentaries and explanations and records discoursed upon by the followers of [Kongzi’s] seventy disciples” (古文
先秦舊書...皆經傳說記，七十子之徒所論。).
583.The only named scholar associated with Liu De in the Hanshu is a Master Mao 毛生, the purported author of
the Mao Shi. See Hanshu 30.1708.
584.Hanshu 30.1712. The Hanshu elsewhere (22.1043) describes Mr. Zhi 制氏 as an expert in music at the
founding of the Han.
585.Cook 1995, p. 5: “It therefore appears that the Yue Ji compiled by Liu De may have been an entirely different
work from our present version of the Yue Ji, and a work which was by Kong Yingda's time in the Tang no longer
extant.”
286
“selected” (cai 采) from earlier texts offers a plausible model for the compilation Lunyu. (Liu
De’s presentation of an “eight-row dance” [ba yi zhi wu 八佾之舞] to Emperor Wu is also intriguing in light of the dance’s mention at Lunyu 3/1, the very first saying in a string of entries
devoted to ritual and music.586) In his sponsorship of scholars and scholarly activities Liu De was
probably aware of a rivalry between himself and other kings, particularly his cousin Liu An 劉
安, a grandson of the Han founder and King of Huainan 淮南 since 164. Although they did not
share the same intellectual predilections—Liu An preferred southern to northern literature, for
instance—both kings must have realized that they competed to a certain extent over the same
pool of intellectual talent.587
The death of Liu De’s father in 141 BCE and the accession of the sixteen year-old Liu Che
劉徹, posthumously known as Emperor Wu 武帝 (r. 141–89 BCE), would have been the most
momentous event of Liu De’s reign. Liu De knew that he would soon be called upon to pay
homage to his newly enthroned half-brother as well as to his grandmother, the Empress Dowager
Dou 竇太后 (d. 135), and the question would have arisen as to the appropriate gifts for the occasion.588 The political implications of this moment would not have escaped him. The imperial
court’s ongoing interest in wresting power away from the regional kings, seven of whom had
launched a rebellion as recently as 154 BCE, meant that the onus was on Liu De to demonstrate
586.Lunyu 3/1: Kongzi said of the Ji clan: “The eight rows of dancers danced in the courtyard. If they can tolerate
this, what can’t they tolerate?” (孔子謂季氏，八佾舞於庭，是可忍也，孰不可忍也？).
587.Compare Vankeerberghen 2001, pp. 9–10: “The competitors for power early in Emperor Wu’s reign also
needed intellectual credentials. Most of them aligned themselves with a text, or with an interpretive tradition, and
many were authors of texts.” In this section I rely heavily on Vankeerberghen’s (2001) discussion of the life and
death of Liu An in the context of early Western Han politics.
588.Hanshu 53.2411. The precise date of Liu De’s court visit is unknown. On Liu An’s court visits, see
Vankeerberghen 2001, pp. 49–51. On court visits generally, which seem to have happened at regular intervals, see
Loewe 2004, p. 372. Empress Dowager Dou was the empress of Emperor Wen 文帝 (180–157 BCE), the mother of
Emperor Jing, and the grandmother of Emperor Wu.
287
his loyalty to Chang’an.589 His gifts to Emperor Wu would also be judged against gifts from other regional kings and that his prestige would depend to a certain extent on their impressiveness
and distinctiveness. Such prestige was a vital form of political capital, without which he would
never have a voice in imperial policy-making or protect himself from persecution in the event
that slanderous officials at Chang’an turned against him.590
Rumors coming out of Chang’an together with Emperor Wu’s early proclamations would
have given Liu De a sense of his newly enthroned half-brother’s interests and predilections.591 He
would have been aware that Emperor Wu had recently issued a call for officials and kings like
himself to recommend virtuous and capable men to Chang’an. At the same time, the emperor had
approved of a proposal to exclude candidates who had “mastered the sayings of Shen Buhai,
Shang Yang, Han Fei, Su Qin, and Zhang Yi” (治申、商、韓非、蘇秦、張儀之言).592 That
589.Vankeerberghen 2001, p. 60: “The Liu kings, until 122 B.C., found themselves in a precarious position. On the
one hand, they were an integral part of how the Han dynasty had defined itself at the outset: whereas as kings they
were entitled to rule over parts of the new empire, as members of the Liu family they could expect special protection
from the emperor when faced with legal prosecution. On the other hand, they were perceived as a threat to the
imperial order: one of them could always rebel to usurp imperial power; moreover, if they engaged in immoral
behavior they could tarnish the reputation of the entire Han ruling family. Thus, while formally the kings were
recognized as corulers, practically they had very little freedom to rule.” See also Bielenstein 1980, pp. 106–107, for
a summary of the punitive bureaucratic reforms (including removal of the privilege to appoint their own high
officials) instituted in 145 BCE following the rebellion in 154.
590.Vankeerberghen’s discussion (2001, pp. 12–13) of Liu An’s attempts to influence Chang’an provides a model
for an activist regional king in the period. See also Major et al 2010, pp. 9–10, for speculation as to Liu An’s
motives for compiling and eventually presenting the Huainanzi to Emperor Wu.
591.Liu De was nominated king of Hejian in 155 and Liu Che king of Jiaodong in 153, so it is likely that the two
princes would have been acquainted with one another at Chang’an. However, extant sources provide no hints into
their personal relationship.
592.Hanshu 6.155–156: “In October of the first year of the Jianyuan reign period [=140 BCE], [Emperor Wu]
issued an edict to the Chancellor, the [Grandee] Secretary, the vassal lords, those with a salary of two-thousand
stones and above, and the vassal lords’ ministers to promote men of accomplishment and integrity who were honest
and forthright and prone to remonstrate. The Chancellor [Wei] Wan memorialized, ‘Of those who have been
recommended, some have mastered the sayings of Shen [Buhai], Shang [Yang], Han Fei, Su Qin, and Zhang Yi,
which throw the governance of the state into chaos. Let all of them be dismissed.’ The memorial was approved.”
建元元年冬十月，詔丞相、御史、列侯、中二千石、二千石、諸侯相舉賢良方正直言極諫之士。丞相綰
奏：所舉賢良，或治申、商、韓非、蘇秦、張儀之言，亂國政，請皆罷。奏可。
288
proposal had been submitted by Wei Wan 衛綰 (d. 130), the current Chancellor (chengxiang 丞
相) and the man who had once served as Liu De’s tutor.593 Recently appointed high officials like
Zhao Wan 趙綰 (d. 139) and Wang Zang 王藏 were also finding Emperor Wu receptive to proposals to redouble the imperial commitment to ancient institutions like the Bright Hall (ming
tang 明堂) and the feng 封 and shan 禪 sacrifices.594 Although the Empress Dowager had succeeded in blocking these efforts for the time being (and also in engineering Zhao Wan’s demise),
she was old and unlikely to live much longer (in fact, she died in 135 BCE).
I imagine that it was for an occasion like this one that Liu De and his Erudites (boshi 博士)
initiated the compilation of a collection of Kongzi’s teachings.595 The larger his library the more
Liu De and his experts would have noticed the remarkable heterogeneity and inconsistency of
Kongzi material across all manner of texts. Annoyingly for those who wished to defend Kongzi’s
reputation, many of these stories and sayings “were not the sayings of a noble man” (非君子之
言), as Mengzi had once put it.596 Liu De might have believed that a collection of Kongzi’s teachings would satisfy a growing interest in Kongzi as a patron sage of the Six Arts (liu yi 六藝).
And if Emperor Wu was interested in purging certain kinds of sayings from imperial discourse
and certain kinds of candidates from consideration for office, what better sayings to champion
than those of Kongzi himself? More selfishly, Liu De might have sought to enhance his credentials as an expert in the ancient models and also showcase the quality and breadth of his library,
the physical manifestation of his commitment to the past and his connection to the ancient sages.
593.Shiji 103.2769.
594.Shiji 28.1384.
595.On the corporate authorship of the Huainanzi at Liu An’s court, which is one possible model for the
compilation of the Lunyu, see Major et al. 2010, pp. 8–9.
596.See the discussion of Mengzi 5A/4 at p. 117.
289
Such an expert would wield considerable influence at a court newly committed to ancient institutions, or so Liu De could have hoped.
Liu De might have also seen this as an opportunity to exploit Kongzi’s reputation as a
paragon of judgment by presenting him as a model of character evaluation for Emperor Wu and
other recommenders across the realm. The assessment and appointment of official candidates
was an especially sensitive issue for Liu De and other kings following Emperor Jing’s decision
in 145 BCE to strip regional kings of the power to appoint their own high officials.597 Perhaps
Liu De had even caught wind of a proposal to establish a semi-formal recruitment system in order to fill the ever-expanding bureaucracy in Chang’an as well as the imperially controlled regional offices.598
Liu De would have assigned the project to a group of favored Erudites, giving them unfettered access to his library and leaving it to them to decide how the project was to be accomplished. The first problem the compilers faced was the question of what to include and what to
leave out. They would have soon realized that each compiler’s ideas about what the text should
look like depended on his own conception of Kongzi. The Chunqiu expert might have wished to
emphasize Kongzi’s authorship of the Chunqiu, the Shi expert his purported involvement in the
editing and compilation of the Shi. Some might have wanted to emphasize Kongzi’s cosmic significance and his powers of prognostication, others his status as an exemplar of the traditional
virtues. And so on.
Several considerations might have pushed the compilers towards a middle-of-the-road, non-
597.Bielenstein 1980, p. 106.
598.This proposal was not accepted until Gongsun Hong 公孫弘 became Chancellor (chengxiang 丞相) in 126. See
Shiji 121.3118.
290
controversial approach to their text. Various powerful officials in Chang’an had studied or were
otherwise partial to particular intellectual or canonical traditions, and the compilers may have
feared that strongly associating their text with one of these traditions over the others would involve them in the intellectual factionalism at court, thereby alienating whole swaths of the Han
elite.599 They avoided associating Kongzi too closely with controversial philosophical positions,
e.g., a theory of human nature (Lunyu 5/13), and they emphasized those aspects of his character
and thought that would appeal to the widest possible audience. In this way, they endeavored to
establish Kongzi as a quotable authority for all. Pronouncements on li 禮 (ritual) and yue 樂 (music), on core values like ren 仁 (humaneness) and xiao 孝 (filial piety), and on legendary figures
like Yao 堯 and Shun 舜 were thought safe, whereas topics like “oddities, physical strength,
chaos, and the spirits” (怪、力、亂、神) (Lunyu 7/21) were deemed too sensitive. Emphasizing
Kongzi’s authorship of the Chunqiu was deemed impolitic given that the text was so strongly associated with political criticism and thus might be used to subvert the moral authority of the Han
imperium. After all, they could have reasoned, better safe than sorry.
Having agreed on the general parameters, the compilers would have then discussed how to
go about putting the text together. Perhaps they resolved first to assemble their own mini-collections of Kongzi material before merging them into a single text. The compilers went their separate ways, each equipped with plenty of ink, brushes, and bamboo strips, as well as his own ideas
599.Vankeerberghen 2001, pp. 9–10: “The competitors for power early in Emperor Wu’s reign also needed
intellectual credentials. Most of them aligned themselves with a text, or with an interpretive tradition, and many
were authors of texts. Dou Ying and Tian Fen reportedly favored classical learning (rushu 儒術); they associated
themselves with the Lu 魯 tradition of Odes (Shi 詩) scholarship by inviting the revered master Shen 申 to court.
Zhufu Yan 主父偃, who enjoyed the emperor’s favor from 128 to 126 B.C., had excelled in studying the arts of the
“Long and Short and Horizontal and Vertical Alliances” (changduan zongheng zhi shu 長短縱橫之術) but then
started studying the Changes (Yi 易), the Spring and Autumn Annals (Chunqiu 春秋), and ‘the words of the hundred
schools.’ The Hanshu’s bibliographical chapter lists a book in twenty-eight chapters by his hand. Ji An is said to
have been adept of the sayings of the Yellow Emperor and of Laozi. Dong Zhongshu and Gongsun Hong 公孫弘
both specialized in the Spring and Autumn Annals.”
291
about Kongzi. A few compilers headed straight to the library to comb the collection for Kongzi
material. Liu De’s copy of the Mengzi was a particularly fruitful source, as were his various Li
禮 texts. One compiler familiar with the Zuozhuan copied an entire Kongzi anecdote (Lunyu
14/21) as well as two statements on ren 仁 (12/1, 12/2). Another wrote out his favorite “Zi yi”
sayings from memory (2/3, 13/22) as well as a dialogue between Kongzi and Zhonggong 仲弓
(13/2) in which he elided those sections which had nothing to do with Liu De’s interest in official recruitment.
Not all of the compilers stuck so closely to available sources of Kongzi stories and sayings.
Some took a more liberal view of what Kongzi yue 孔子曰 meant and chose to interpret it as
“Kongzi might very well have said” or “Kongzi should have said.” For them, a proper Kongzi
saying derived its authority from its recognizability, not its textual pedigree. These compilers introduced popular proverbs (e.g., 7/1) or added their favorite sayings from texts not attributed to
Kongzi. One ritual expert copied down a set of ritual prescriptions to which he affixed the word
“Kongzi 孔子,” thereby turning it into a description of Kongzi’s conduct (Lunyu book 10).
Another compiler with a predilection for lists devised a series of capsule descriptions of
Kongzi’s conduct and teachings (7/4, 7/13, 7/18, 7/21, 7/25).
When the compilers reconvened to share their individual collections with the group, they
might have been daunted by the bundles of bamboo strips that confronted them, each of which
represented its compiler’s distinct conception of Kongzi. Looking at this material, they would
have realized the need to condense and standardize it in some way if it was to be of any use.
Here they would have remembered that they were working for an audience of one, Liu De, and
that Liu De intended to present the text to Emperor Wu. For their king and emperor as well as for
the officials responsible for drafting official documents (Lunyu 14/8), “useful” meant “quotable.”
292
Thus they might have decided to isolate the most important Kongzi sayings from longer Kongzi
dialogues and anecdotes, preferably using no more than a single bamboo strip per entry. In addition to enhancing the text’s quotability, this was also the simplest way to put the raw source materials into a more or less consistent format. However, some compilers might have objected that
certain dialogues could not be compressed into single sayings and insisted that these texts were
important enough to be included in their original format. Among these were the interview scenes
(Lunyu 5/26 & 11/26) in which Kongzi assessed his disciples’ qualifications for office, episodes
which addressed Liu De’s concern to portray Kongzi as an exemplary judge of talent and
character.
After paring and pruning their mini-collections, the compilers still would have faced the
problem of merging them into a coherent text. The solution they seem to have hit upon was to
group individual entries together according to shared themes. Strips with sayings having to do
with ritual were deposited in one pile, Kongzi’s assessments of his disciples in another, assessments of non-disciples in another, and so on.600 If Wang Chong’s mention of a Hejian Lunyu in
nine bundles (juan 卷) is any indication,601 Liu De’s compilers might have ended up with a total
of nine groupings:
theme
location in the received Lunyu
the virtues (ren 仁, xiao 孝,
dao 道)
books 2 & 4, 6/18–23
ritual and music
2/23–24 and the whole of book 3
assessing disciples
5/1–5/14, 5/25–5/28, 6/1–6/14, 9/20–22, book 11
600.Vankeerberghen’s (2001, p. 5) description of the Huainanzi also fits the received Lunyu: “As the Huainanzi
resulted from a cooperative effort by several authors, there is very little unity at the level of terminology. Words
mean different things and are evaluated differently from chapter to chapter or from passage to passage within a
given chapter. It is also hard to find unity at the level of ideas. Ideas overlap or even contradict one another if taken
at face value.”
601.Appendix 2:F.
293
assessing non-disciples
5/15–5/21, 5/23–24, 8/1, 8/11, 8/18–21, 14/12–19
Kongzi’s self-assessments
book 7, 9/1–18
Kongzi’s ritual conduct
book 10
governance
book 12, 13/1–20
disciple sayings
1/2, 1/4, 1/7, 1/9, 1/10, 1/12, 1/13, book 19
miscellaneous sayings
14/1–11, 14/20–44, 15, 16, & 17
Sayings within these groupings were then grouped together with other sayings that used similar
language or that were about similar topics (e.g., the string of junzi descriptions at 15/18–23). In
one section, all of the sayings about fathers and/or mothers (fumu 父母) were placed in a sequence (4/18–21); in the section on Kongzi self-assessments, three sayings using the character qi
齊 and its written variant zhai 齋 were listed in a row (7/13–15).602 The compilers might have
also given some consideration to which entries belonged at the beginning of each section. For instance, in the ritual and music subsection, they might have put Kongzi’s statement on the ba yi
八佾 (“eight rows of dancers”) first as a nod to their patron’s interest in ancient music.
At this point they would have ordered a scribe to copy out a first draft of their compilation,
which they would have promptly presented to Liu De. Liu De might have drawn up a list of revisions that included a few additions of his own as well as suggestions for which sayings should go
where. Being particularly sensitive to the political implications of the work, he might have insisted that the compilers edit an episode from the Zuozhuan that featured Kongzi disobeying a direct
command from his lord (14/21) and that they remove all of Kongzi’s criticisms of late Spring
and Autumn-era rulers lest Emperor Wu understand them as indirect criticisms of his own rule.
The compilers would have followed their patron’s instructions to the letter and returned with a fi-
602.Simson 2006, p. 48. Simson (pp. 45–50) identifies a number of other such clusters.
294
nal version of the text.
When Liu De received the imperial summons to travel to Chang’an, he would have presented
his collection along with a copy of the “Yue ji” and a performance of the ancient ba yi 八佾
(eight row) dance. Liu De would have given a lot of thought to the question of how to introduce
his text to Emperor Wu. Rather than describe the actual circumstances of its creation, he may
have explained that his collection had been recovered from certain “ancient texts” (gu wen 古文)
submitted by Ru from Kongzi’s ancestral region in Lu. These Ru, he could have said, had received Kongzi’s teachings via generations of masters and disciples down to the present day.
Although Emperor Wu was too sophisticated a consumer of “ancient texts” to accept Liu
De’s account, he would have appreciated Liu De’s humility and the appeal of his text. At a time
when more and more scholars were claiming Kongzi as their intellectual progenitor, especially
the soon-to-be appointed “Erudites of the Five Classics” (wu jing boshi 五經博士),603 Liu De’s
text offered unmediated access to the creator of the canon and the patron saint of traditional
learning. No longer would the emperor have to rely on scholars to articulate the wisdom of the
(often impenetrable) classics for him. He might have also appreciated the quotability of Liu De’s
text, which stood in stark contrast to the poetic language of the Shi, the archaic language of the
Shu, the mysterious images of the Yi, or the coded language of the Chunqiu. Most of all, he
would have seen that Liu De’s Kongzi was a figure who derived his authority exclusively from
his learning and wisdom; unlike the Chunqiu Kongzi, he was not a paragon of political critique
who could inspire dissenting voices.604
At first only a small circle of Emperor Wu’s advisers would have known of and had access to
603.According to the Hanshu (6.159), these positions were not appointed until 136 BCE.
604.On Kongzi’s association with the Chunqiu in the Han, see Kern 2011.
295
the text. However, copies might have begun circulating out of Liu De’s court in Hejian and
spreading throughout the Qi and Lu regions. Such texts might have been the direct ancestors of
the Qi Lunyu and Lu Lunyu listed in the “Yiwen zhi.” (Perhaps it was one of these copies which
Sima Qian had obtained for the compilation of the “Kongzi shijia.”) When Emperor Wu’s son
and designated crown prince Liu Ju 劉據 (129–91) had reached an appropriate age, Emperor Wu
ordered one of the boy’s tutors to instruct him in the text.605 The same tutor might have also suggested certain revisions to enhance its pedagogical value. These included adding a reworked
book one which emphasized “learning” (xue 學) and “filial piety” (xiao 孝) (e.g., 1/1, 1/2), some
material from Sima Qian’s recently finished biography of Kongzi (18/3–8), and a number of
Zengzi sayings meant to establish Zengzi’s authority as author of the Xiaojing, another text being
used as a textbook for princes at Chang’an. At this point the Lunyu might have been reorganized
into twenty sections (perhaps splitting each of Liu De’s nine sections into two and adding books
one and twenty) in order to make individual books shorter and thus easier to digest.
When news of the crown prince’s curriculum spread, there would have been a scramble within the Liu 劉 clan to secure copies of the text for other eligible princes. Quotations of the text
soon began to appear in memorials, and before long mastery of the Lunyu was perceived as a
pre-requisite for selection to higher office. Self-proclaimed experts in the Lunyu soon emerged to
take advantage of the text’s popularity, offering instruction in the text to the highest bidder. By
the time of Emperor Wu’s death and the succession of his teenage son Liu Fuling 劉弗陵, the future Emperor Zhao, the Western Han elite were not in the least bit surprised to hear that Emperor
Zhao was studying the Lunyu.
605.The Hanshu (63.2741) records that Liu Ju received instruction in the Gongyang 公羊 and Guliang 穀梁
commentaries to the Chunqiu, but it is highly unlikely that these were the only texts he was exposed to.
296
*
*
*
This is by no means the only scenario which might account for the Lunyu’s compilation in the
Western Han. Perhaps the text was put together under the auspices of another regional king, perhaps even the very King Gong of Lu 魯恭王 whom legend credits with having been indirectly
responsible for the discovery of the “ancient script” (guwen 古文) Lunyu.606 Or perhaps the Lunyu owes its existence to figures at Chang’an 長安 working under Emperor Wu or his father, Emperor Jing, or his grandfather, Emperor Wen. Someone like Kong Anguo 孔安國, a supposed descendant of Kongzi in the twelfth generation who served as Grandee Remonstrant (jian dafu 諫
大夫) during Emperor Wu’s reign, might have had the motivation, education, textual resources,
and prestige to compile and market a compilation of his ancestor’s sayings to a wider audience.607 It is also conceivable that the Lunyu was first compiled by people outside of a court context, perhaps by self-identifying Ru 儒 who presented the text to Chang’an or to a regional king
like Liu De. But whatever the circumstances of its creation, I believe students of the Lunyu have
much to gain from thinking of it as a product of a Western Han intellectual, textual, and political
milieu.
606.For my discussion of King Gong of Lu, see p. 208.
607.Hanshu 88.3607. The Hanshu “Yiwen zhi” (30.1706.) records that Kong Anguo came into the possession of the
ancient script texts discovered by King Gong, including the ancient text Lunyu.
297
SAYINGS OF CONFUCIUS, DESELECTED
Volume 2
Michael Justin Hunter
A DISSERTATION
PRESENTED TO THE FACULTY
OF PRINCETON UNIVERSITY
IN CANDIDACY FOR THE DEGREE
OF DOCTOR OF PHILOSOPHY
RECOMMENDED FOR ACCEPTANCE
BY THE DEPARTMENT OF
EAST ASIAN STUDIES
Adviser: Willard Peterson
September 2012
CHAPTER ONE APPENDICES
• Appendix A: “Between Chen and Cai” stories through the Western Han
Mengzi 7B/18 (SBCK When the noble man [=Kongzi] was trapped between Chen and Cai... (君
14/6b)
子之厄於陳蔡之間...)
Mozi 39 (SBCK 9/22a; When Kong So-and-so was at rock bottom between Chen and Cai he ate
KZJY 13.31/p. 289)
only simple broth without any grain... (孔某窮於蔡陳之閒，藜羹不糝)
Lüshi chunqiu 14/6
(SBCK 14/17a–b; KZJY
13.28/p. 287)
Lüshi chunqiu 17/3
(SBCK 17/9b–10a; KZJY
13.32/p. 289)
Zhuangzi 20 (SBCK
7/20b; KZJY 17.7/p. 388)
Zhuangzi 20 (SBCK
7/24b; KZJY 13.30/pp.
288–89)
Zhuangzi 28 (SBCK
9/27b; KZJY 13.27/pp.
286–87)
When Kongzi was at rock bottom between Chen and Cai for seven days
he did not eat anything except simple broth without any grain... (孔子窮
於陳、蔡之間，七日不嘗食，藜羹不糝。)
When Kongzi was at rock bottom between Chen and Cai, even simple
broth was not poured and for seven days he did not eat grain. He slept
during the day. (孔子窮乎陳、蔡之間，藜羹不斟，七日不嘗粒，晝
寢。)
When Kongzi was surrounded between Chen and Cai for seven days he
had no food to cook. (孔子圍於陳蔡之間，七日不火食。)
When Kongzi was at rock bottom between Chen and Cai for seven days
he had no food to cook. (孔子窮於陳蔡之間，七日不火食。)
When Kongzi was at rock bottom between Chen and Cai for seven days
he had no food to cook. He ate only simple broth with no grain, he looked
exhausted, yet he strummed and sang in his room. (孔子窮於陳蔡之間，
七日不火食，藜羹不糝，顏色甚憊，而弦歌於室。)
Huainanzi 16 (SBCK To be Kongzi and to abandon the Six Arts while at rock bottom between
16/17a)
Chen and Cai would have been [the height of] delusion. (為孔子之窮於
陳、蔡而廢六藝，則惑。)
Hanshi waizhuan 7
(SBCK 7/3b; KZJY
13.24/pp. 283–84)
When Kongzi was in trouble between Chen and Cai he sat on the “threeclassics mat” and did not eat for seven days except for simple broth with
no grain. Though his disciples looked famished he read aloud from his
books and practiced ritual and music without stopping. (孔子困於陳蔡之
間，即三經之席，七日不食，藜羹不糝，弟子有飢色，讀書習禮樂
不休。)
Shiji 47.1930
[Kongzi while stuck between Chen and Cai] could not proceed, his food
ran out, his followers were exhausted, and no one could rise...” (不得
行，絕糧。從者病，莫能興。).
Shuiyuan 17 (SBCK
When Kongzi encountered difficulties at the border between Chen and
Cai his food ran out and his disciples all looked famished... (孔子遭難
陳、蔡之境，絕糧，弟子皆有饑色...)
17/10a–b; KZJY 13.26/
pp. 285–86)
298
Shuiyuan 17
(SBCK 17/11b–12a;
KZJY 13.25/pp. 284–85)
When Kongzi was in trouble between Chen and Cai he was hemmed in
on all sides. He sat on his “three-classics mat” and did not eat for seven
days except for simple broth with no grain. Though his disciples all
looked famished he recited the Odes and Documents and practiced ritual
without stopping. (孔子困於陳、蔡之間，居環堵之內，席三經之席，
七日不食，藜羹不糝，弟子皆有饑色，讀詩書治禮不休。)
299
• Appendix B: Passages in which Kongzi enjoins disciples to “take note of” (ji zhi 記之/zhi zhi
志之/shi zhi 識之) or “work on” (mian zhi 勉之) his words
ji zhi 記之
Shuiyuan 7, “Zheng Kongzi said, “Disciples, take note: Duke Huan was an overlord and Guan
li 政理” (SBCK 7/5b; Zhong an accomplished assistant, yet they still took wisdom as foolishKZJY 10.57/p. 232)
ness. [If even they could make such a mistake,] how much more those
who are inferior to Duke Huan and Guan Zhong?”
孔子曰：弟子記之：桓公，霸君也；管仲，賢佐也；猶有以智為愚
者也，況不及桓公管仲者也。
Shuiyuan 10, “Jing
shen 敬慎” (SBCK
17a–b; KZJY 12.9/p.
255)
Kongzi turned to his disciples and said, “Take note: although vulgar, this
saying goes to the heart of the matter. A Shi says, ‘Tremblingly, circumspectly, as if overlooking a deep chasm or treading on thin ice.’ If you
carry yourself like this, how could you bring disaster on yourself with
your mouth?”
孔子顧謂弟子曰：記之：此言雖鄙，而中事情。詩曰：戰戰兢兢，
如臨深淵，如履薄冰。行身如此，豈以口遇禍哉！
Shuiyuan 10, “Jing
shen 敬慎” (SBCK
Kongzi said, “Disciples, take note: let this serve as a warning to you.”
Thereupon thirteen of his disciples returned home to care for their
10/18b; KZJY 2.6/p. 16) parents.
孔子曰：弟子記之，此足以為戒也。於是弟子歸養親者十三人。
Shuiyuan 17, “Za
yan 雜言” (SBCK
Kongzi said, “You [=Zilu], take note, because I will tell you: those who
show off their speech are grandiloquent, and those who show off their
17/17a; KZJY 3.18/p. 33) conduct are arrogant...”
孔子曰：由，記之，吾語若：賁於言者，華也，奮於行者，伐
也。...
Xinxu 5, “Za shi 雜 Kongzi turned to Zigong and said, “Disciples, take note: when a govern事” (SBCK 5/8a; KZJY ment is unfair and its officers are too harsh, it is worse than tigers and
10.54/p. 231)
wolves.”
孔子顧子貢曰：“弟子記之。夫政之不平而吏苛，乃甚於虎狼矣。
Kong congzi 6, “Jie Kongzi learned of this and said, “Followers, take note: Yanzi served
Mo 詰墨” (SBCK
three lords with one heart. He is a [true] junzi.”
6/5a)
孔子聞之，曰：小子記之。晏子以一心事三君，君子也。
zhi zhi 志之
300
Guoyu, “Lu yu xia Zhongni learned of this and said, “Disciples, take note: this lady of the Ji
魯語下” (SBCK 5/12a; clan was not licentious.”
KZJY 9.30/p. 179)
仲尼聞之曰：弟子志之，季氏之婦不淫矣。
Xunzi 29, “Zi dao 子 Kongzi said, “You [=Zilu], take note, because I will tell you: even if you
道” (SBCK 28/11a–11b; possess the might of a soldier, you cannot raise yourself up. It is not beKZJY 2.13/pp. 18–19)
cause you lack strength, but because the situation does not allow it...”
孔子曰：由志之，吾語女。雖有國士之力，不能自舉其身，非無力
也，勢不可也。...
Xunzi 29, “Zi dao 子 Kongzi said, “You [=Zilu], take note, because I will tell you: those who
道” (SBCK 29/12b;
show off their speech are grandiloquent, and those who show off their
KZJY 3.16/p. 32)
conduct are arrogant...”
孔子曰：由志之！吾語汝：奮於言者華，奮於行者伐。...
Hanshi waizhuan 3
(SBCK 3/20b; KZJY
3.17/pp. 32–33)
Kongzi said, “You [=Zilu], take note, because I will tell you: those who
show off their speech are grandiloquent, and those who show off their
conduct are arrogant...”
孔子曰：由志之，吾語女；夫慎於言者不譁，慎於行者不伐。
Kongzi jiayu 9, “San The Master said, “You [=Zilu], take note, because I will tell you: those
shu 三恕” (SBCK
who show off their speech are grandiloquent, and those who show off
2/16a)
their conduct are arrogant...”
子曰：由志之！吾告汝！奮於言者華，奮於行者伐。...
Kongzi jiayu 10,
“Hao sheng 好生”
Kongzi heard this saying and said, “Followers, take note: who said Can
[=Zengzi] does not understand ritual?”
(SBCK 2/16a)
孔子聞斯言也，曰：二三子志之！孰為參也不知禮也？
Kongzi jiayu 22,
“Kun shi 困誓”
(SBCK 5/21b)
Kongzi said, “You [=Zilu], take note, because I will tell you: even if you
possess the strength of a soldier, you cannot raise yourself up. It is not
because you lack strength, but because the situation does not allow it...”
孔子曰：由！汝志之，吾語汝。雖有國士之力，而不能自舉其身，
非力之少，勢不可矣。...
Kongzi jiayu 41,
“Zheng lun 政論”
Kongzi learned of this and said, “Disciples, take note: This lady of the Ji
clan can be said to have ‘not transgressed.’”
(SBCK 9/28a)
孔子聞之，曰：弟子志之！季氏之婦，可謂不過矣。
shi zhi 識之
301
Liji, “Tan gong
shang 檀弓上”
(SBCK 2/10a; KZJY p.
520)
When Kongzi was in Wei there was a funeral party, and when the Master
observed it he said, “What a fine funeral! It can serve as a model—disciples, take note!...” The Master said, “Disciples take note: I am unable to
put this into practice.”
孔子在衛，有送葬者，而夫子觀之，曰：善哉為喪乎！足以為法
矣。小子識之。...子曰：小子識之：我未之能行也。
Liji, “Tan gong xia The Master said, “Disciples, take note: a cruel government is fiercer than
檀弓下” (SBCK 3/13b; a tiger.”
KZJY p. 524)
夫子曰：小子識之：苛政猛於虎也。
Yanzi chunqiu ,
“Wen xia 問下”
Zhongni learned of this and said, “Disciples, take note: Yanzi served a
hundred lords with one heart.”
(SBCK 4/20b; KZJY 7.8/
p. 148)
仲尼聞之曰：小子識之！晏子以一心事百君者也。
Zhuangzi 5, “De
chong fu 德充符”
(SBCK 2/35b; KZJY
16.26/p. 374)
mian zhi 勉之
Kongzi said, “Disciples, work on this: this toeless cripple still endeavors
to learn and improve upon his former misconduct. How much more
should those of flawless virtue?”
孔子曰：弟子勉之！夫無趾，兀者也，猶務學以復補前行之惡，而
況全德之人乎！
Zhonglun , “Zhi xue Kongzi said, “If you refuse to learn how will you conduct yourself? If
治學” (SBCK 上/2b; you refuse to think how will you obtain [what you want]? Disciples,
KZJY 1.28/p. 10)
work on this: this can serve as a teacher for others.”
孔子曰：弗學何以行？弗思何以得？小子勉之，斯可以為人師矣。
Zhonglun, “Xiu ben Kongzi said, “Disciples, work on this: do not discard yourself. Others
修本” (SBCK 上/10a; will discard you no matter what, so why discard yourself? How distantly
KZJY 1.29/p. 10)
others will keep away from you.”
孔子曰：弟子勉之！汝毋自舍，人猶舍汝，況自舍乎！人違汝其遠
矣。
302
• Appendix C: The Shanghai Museum “Lu bang da han” manuscript
There was a great drought in Lu. Duke Ai said to Kongzi, “Master, will you not divine its cause
for us?”
Kongzi replied, “When there is a great drought, is it not because [the state] has lost punishments and virtue? Simply rectify punishments and virtue.”
Duke Ai said, “But the masses will think that I do not know to serve the spirits with the shuo
rain sacrifice.608 What is to be done about that?”
Kongzi said, “The masses understand that the shuo sacrifice serves the spirits, but they do not
understand punishments and virtue. If you do not begrudge609 precious jades and silks to the
mountains and rivers [and also] rectify punishments and virtue and thereby serve Heaven on
high, then the spirits will respond and the great drought will certainly come to an end.
The Duke said, “Excellent!”
Kongzi exited and ran into Zigong, saying, “Zigong! You have heard the word on the street.
Do they not say that my replies were wrong?”
Zigong said, “To the contrary, [they say that] my Master values supplication. Rectifying punishments and virtue and thereby serving Heaven on high—this is the right thing to do. But not
begrudging precious jades and silks to the mountains and rivers—isn’t that impermissible? The
mountains have the rocks for their skin and trees for their people. If it does not rain the stones become scorched and the trees die, so the mountains’ desire for rain far exceeds our own. Why
must they rely on our supplications? The rivers have water for their skin and fish for their people.
If it does not rain the water dries up and the fish die, so the rivers’ desire for rain far exceeds our
own. Why must they rely on our supplications?”
Kongzi said, “Ah, Ci [=Zigong]! I will tell you: a noble man considers supplication to be a
matter of patterning, whereas the masses consider it a matter of the spirits. If he does not supplicate, how could a king or duke eat his fill of grains and meats without losing control over the
masses?”
魯邦大旱，哀公謂孔子：子不為我圖之？
孔子對曰：邦大旱，毋乃失諸刑與德乎？唯〔正刑與德。
哀公曰：庶民以我不知以說之事鬼也。若〕之何哉？
孔子曰：庶民知說之事鬼也，不知刑與德。汝毋愛珪璧幣帛於山川，正刑與
608.For a discussion of this sacrifice, see Shanghai bowuguan cang Zhanguo Chu zhushu, vol. 2, p. 206.
609.Ma Chengyuan (Shanghai bowuguan cang Zhanguo Chu zhushu, vol. 2, p. 206.) reads ai 愛 (begrudge) as ai 薆
(hide or bury), which is functionally equivalent to yi 瘞, a character which appears in the Mao Shi (#258) in a similar
context. However, Li Xueqin cites a parallel from the same Shi to support the ai 愛 reading, which seems more
natural given that Kongzi recommends rectifying punishments and virtue in addition to conducting sacrifices.
303
〔德〕，〔以事上天，鬼神感之，大旱必止矣。
公曰：善哉。
孔子〕出，遇子貢曰：賜爾聞巷路之言，毋乃謂丘之對非歟？
子貢曰：否，抑吾子乃重命其歟？若夫正刑與德，以事上天，此是哉。若夫毋愛珪璧
幣帛於山川，毋乃不可。夫山，石以為膚，木以為民，如天不雨，石將焦，木將死，其欲
雨有甚於我，又必恃乎命乎？夫川，水以為膚，魚以為民，如天不雨，水將涸，魚將死，
其欲雨有甚於我，何必恃乎命乎？
孔子曰：鳴呼！〔賜也，我告爾：命也者，君子以為文，庶民以為神。如不命，王〕
公豈不飽粱食肉哉，抑無如庶民何？610
610.Here I follow Liao Mingchun’s (2004) proposed reconstruction with the supplementary material in brackets
〔〕.
304
• Appendix D: Zigao debunks Kongzi rumors in the Kong congzi 孔叢子
Prince Pingyuan was drinking with Zigao and pressed him to drink more wine, saying, “As the
old proverbs have it, ‘Yao and Shun could drink a thousand jars of wine and Kongzi a hundred
beakers. Zilu drank in small sips, but he could still drain ten goblets. That is to say, among the
sages and the accomplished men of old there was none who could not drink. Why are you reluctant to do so?” Zigao answered, “According to what I have heard, it was by means of the Way
and virtue that accomplished men and sages surpassed others, not through food and drink.”
Prince Pingyuan said, “If what you say is right, then how did these sayings arise?” Zigao answered, “These sayings arise from people who love to drink, and the sayings are no doubt inciting and provocative words. They are not true.” Prince Pingyuan was pleased and said, “Had I not
provoked you, I would not have heard these elegant words of yours.”
平原君與子高飲，強子高酒曰：昔有遺諺，堯舜千鍾，孔子百觚，子路嗑嗑。尚飲十榼。
古之聖賢無不能飲也。吾子何辭焉？子高曰：以穿所聞賢聖以道德兼人，未聞以飲食也。
平原君曰：即如先生所言，則此言何生？子高曰：生於嗜酒者，蓋其勸厲獎戲之辭，非實
然也。平原君欣然曰：吾不戲子無所聞此雅言也。611
*
*
*
Prince Pingyuan said to Zigao, “I have heard that your forefather had an intimate interview with
Nanzi, the wife of the Duke of Wei. I have also heard that when your forefather was traveling
south he passed through Agu, where he exchanged words with a washerwoman. Did this actually
occur?” Zigao answered, “Men who trust one another do not believe gossip when they hear it.
Why? Because they view the case in light of the past activities of those to whom the story is ascribed. Formerly, during my forefather’s sojourn in Wei, the Prince of Wei asked him about
commanding the troops. But my forefather dismissed the question without statement, and though
the question remained unsettled he seized the reins of his carriage and departed. Since the Prince
of Wei’s request for an interview was apparently not carried out, how could his wife have been
able to meet intimately with my forefather? It seems that in the remote past, even when the rites
and etiquette were on the decline, there were those who put into effect the rule concerning the attendance of women in the great sacrificial feast. To suggest that the wife of the Prince of Wei
conducted such a feast with the Master is to think of the Master as not having grasped [the rites].
As for his conversation with the washerwoman in Agu, this [story] arose in recent times. It is the
work of someone who made it up in order to further his own motives.
611.Kong congzi 13 (SBCK 4/8a–b). Translation adapted from Ariel 1989, p. 136.
305
平原君問子高曰：吾聞子之先君，親見衛夫人南子，又云南遊遇乎阿谷而交辭於漂女。信
有之乎？答曰：士之相信，聞流言而不信者。何哉？以其所巳行之事占之也。昔先君在
衛，衛君問軍旅焉。拒而不告，問不巳攝駕而去。衛君請見，猶不能終。何夫人之能覿
乎？古者大饗，夫人與焉。於時禮儀雖廢，猶有行之者。意衛君夫人饗夫子則夫子亦弗獲
巳矣。若夫阿谷之言，起於近世，殆是假其類以行其心者之為也。612
612.Kong congzi 13 (SBCK 4/8b–9a). Translations adapted from Ariel 1989, pp. 136–37.
306
• Appendix E: Mengzi debunks Kongzi rumors at Mengzi 5A/8
Wan Zhang asked, “Some say that Kongzi stayed with an ulcer doctor in Wei, and he stayed with
the servant Qi Huan in Qi. Did he?”
Mengzi said, “No, he did not. [Such sayings] are the work of those who like telling tales. He
stayed with Yan Chouyou in Wei. The wives of Mizi and Zilu were siblings, and Mizi said to
Zilu, “If Kongzi stays with me, he could became a minister of Wei.” When Zilu informed
Kongzi of this, he said, “If it has been mandated.” Kongzi advanced according to ritual and retired according to what was proper. Whether he obtained something or not, he would say, “It has
been mandated.” Were he to have stayed with an ulcer doctor or the servant Qi Huan, that would
have been neither proper nor mandated. After Kongzi was displeased with Lu and Wei, he ran
into Master of Horses Huan in Song, who sought to capture and kill him. He passed through
Song by traveling incognito. At that time Kongzi was in dire straits and he stayed with Master of
the City Zhenzi, who was the Marquis of Chen and a subject of Zhou. I have heard that to observe a nearby subject one considers his guest; to observe a distant subject one considers his
host. If Kongzi had stayed with an ulcer doctor and the servant Qi Huan, how could he have been
Kongzi?
萬章問曰：或謂孔子於衛主癰疽，於齊主侍人瘠環，有諸乎？孟子曰：否，不然也；好事
者為之也。於衛主顏讎由。彌子之妻與子路之妻，兄弟也。彌子謂子路曰：孔子主我，衛
卿可得也。子路以告。孔子曰：有命。孔子進以禮，退以義，得之、不得、曰有命。而主
癰疽與侍人瘠環，是無義無命也。孔子不悅於魯衛，遭宋桓司馬，將要而殺之，微服而過
宋。是時孔子當阨，主司城貞子，為陳侯周臣。吾聞觀近臣，以其所為主；觀遠臣，以其
所主。若孔子主癰疽與侍人瘠環，何以為孔子？613
613.SBCK 9/14b–15a; KZJY p. 514.
307
CHAPTER TWO APPENDICES
• Appendix A: Lunyu 13/18 parallels
Lüshi chunqiu 11/4.3 (SBCK 11/8a–b;
Hanshi waizhuan
Baihu tongyi
KZJY 2.17/p. 20)
(SBCK 2/9b)
(SBCK 4/12b)
Lunyu 13/18
The Duke of She said to
Kongzi,
In Chu there was one Upright Self who informed the authorities after his father stole a
sheep. The authorities had his father arrested
and were about to execute him when Upright
Self asked to take his place. Upright Self was
about to be executed when he said to the official, “Is it not honest to report one’s father
when he steals a sheep? Is it not filial to ask
to replace one’s father when he is about to be
executed? If a state would execute an honest
and filial man, then who wouldn’t it execute?” The king of Chu heard this and pardoned him.
Kongzi said,
Kongzi learned of this and said,
“Sons cover for their
fathers and fathers
“Upright Self’s honesty is strange indeed! He cover for their sons.
took advantage of his own father to make a
Therein lies
name for himself.” Thus Upright Self’s hon- uprightness.”
esty was worse than being dishonest.614
楚有直躬者，其父竊羊而謁之上，上
執而將誅之。直躬者請代之。將誅
矣，告吏曰：父竊羊而謁之，不亦信
乎？父誅而代之，不亦孝乎？信且孝
而誅之，國將有不誅者乎？荊王聞
之，乃不誅也。
孔子聞之曰：
異哉直躬之為信也，一父而載取名
焉。
故直躬之信，不若無信。
“Among my people
there is a man named
Upright Self who testified against his father
after he stole a sheep.”
The Lunyu says,
Kongzi said,
“Fathers cover for
their sons and
sons cover for
their fathers.
Therein lies
uprightness.
“The upright ones
among my people are
different from this man.
Fathers cover for their
sons and sons cover for
their fathers. Therein
lies uprightness.”
葉公語孔子曰：吾黨
有直躬者，其父攘
羊，而子證之。
論語曰：
孔子曰：
子為父隱，父為子 父為子隱，子
隱，直在其中矣。 為父隱，直在
其中矣。
孔子曰：
吾黨之直者異於是：
父為子隱，子為父
隱。直在其中矣。
614.Here we have a good example of an ambiguous quotation ending. In the Knoblock and Riegel edition (2000, pp.
251–52), the Chinese text concludes the Kongzi attribution after the line “He took advantage of his own father to
make a name for himself,” but their English translation concludes it after “Thus Upright Self’s honesty was worse
than being dishonest.” I have opted for the former punctuation on the grounds that gu 故 marks the Lüshi chunqiu
compiler’s own comment on the anecdote.
308
• Appendix B: Kongzi attributions in Eastern Han edicts
ruler
year
source quoted
quotation marker
Cao Cao 曹操 (155–
220)615
216
214
212
Lunyu 9/3
Lunyu 5/17
Lunyu 5/9
Zhongni yue 仲尼曰 (Sanguo zhi 1.47)
Kongzi cheng 孔子稱 (Sanguo zhi 10.325)
Zhongni you yan 仲尼有言 (Sanguo zhi
9.271–72)
211
Lunyu 8/21
Kongzi suo wei 孔子所謂 (Sanguo zhi
16.496)
Xian 獻帝 (r. 189–220)
210
Lunyu 8/20
Lunyu yun 論語云 (Sanguo zhi 1.32–33)
207
Zuozhuan
Zhongni yunti 仲尼隕涕 (“Zhongni shed
tears”) (Sanguo zhi 22.650)
195
Lunyu 7/11
193
Lunyu 7/3
Kongzi shen 孔子慎 (“Kongzi was cautious...”) (Hou Han ji jiaozhu 28.794)
Kongzi tan 孔子歎 (Hou Han ji jiaozhu
27.766)
Empress Dowager Deng
鄧 (b. 81–d. 121)
106
106
Lunyu 8/20
Lunyu 8/21
Kongzi yue 孔子曰 (Hou Hanshu 76.2469)
Kongzi yue 孔子曰 (Hou Hanshu 4.197)
He 和帝 (r. 88–106)
?
Lunyu 4/3
Kongzi yue 孔子曰 (Hou Hanshu 50.1675)
Zhang 章帝 (r. 75–88)
80
79
Lunyu 13/3
Kongzi yue 孔子曰 (Hou Hanshu 3.140)
Lunyu 7/3, 19/6616 Kongzi yue 孔子曰 (Hou Hanshu 3.138)
75
Lunyu 6/14, 13/2
Kongzi hui 孔子誨 (“Kongzi taught”)
(Hou Hanshu 3.133)
Ming 明帝 (r. 57–75)
69
Guangwu 光武帝 (r. 25– 56
57 CE)
56
Lunyu 11/8
Zhongni zang zi 仲尼葬子 (“Zhongni
buried his son”) (Hou Hanshu 2.115)
Shiji (?)617
Kongzi yue 孔子曰 (Hou Hanshu 16.601–
02)
Lunyu 3/6
Zhongni fei 仲尼非...
(“Zhongni criticized...”) (Hou Hanshu
97.3164)
52
26
Lunyu 16/1
Lunyu 13/3
Kongzi yue 孔子曰 (Hou Hanshu 18.695)
Kongzi yun 孔子云 (Hou Hanshu 1a.29)
615.Although not officially an emperor, Cao Cao issued a number of ling 令 (“commands”) as the effective ruler of
what eventually became the state of Wei 魏 under his son Cao Pi. This list only includes those ling listed in the
Sanguo zhi and excludes other fragments collected by Yan Kejun in the Quan Sanguo wen.
616.The first part of the quotation reads, “Kongzi said, ‘To learn without progressing—this is my worry’” (孔子
曰：學之不進，是吾憂也。), which is probably a variant of Lunyu 7/3 which substitutes jin 進 for jiang 講.
617.Shiji 67.2188: “Kongzi said, ‘Since I have had [Yan] Hui, my followers have become ever closer” (孔子曰：自
309
• Appendix C: Kongzi attributions in Eastern Han commentaries
Text618
Explicit Lunyu
attributions/
references
He Xiu’s 何休 (129–182) 1
Gongyang 公羊
commentary
Total # of #/% from
Kongzi
Lunyu
quotations
36
27 (75%)
Zhao Qi’s 趙岐 (d. 201)
Mengzi commentary
14
1
1 (100%)
Zheng Xuan 鄭玄’s
(127–200) Liji
commentary
10
12
12 (100%)
Zheng Xuan Mao Shi 毛 2
詩 commentary
4
3 (75%)
Zuozhuan (x1);
Zheng Xuan Yili 儀禮
commentary
7
6 (86%)
Liji (x1)
Zheng Xuan Zhouli 周禮 17
commentary
9
4 (44%)
Liji (x4); unknown
(x1)
Xu Shen 許慎 (c. 55–c.
149)/Gao You 高誘 (c.
168–212) Huainanzi
commentary619
14
5
4 (80%)
Liji (x1)
Gao You’s Lüshi chunqiu 22
commentary
8
7 (88%)
Xiaojing (x1)
Wang Yi’s 王逸 (fl. c.
120 CE) Chuci 楚辭
commentary
0
0
3
11
Other
Liji (x3); Zuozhuan
(x1); Shiji 47 (x1);
unknown (x4)
吾有回，門人益親。). The same saying also appears at Shangshu dazhuan (SBCK 2/19a; KZJY 7.1/p. 146), Kongzi
jiayu (SBCK 9/1a), and Kong congzi (SBCK 1/10b), but not in the Lunyu.
618.I have used the Sibu congkan editions for all of the commentaries listed here.
619.See Loewe (ed.) 1993, pp. 190–191, for the history of these commentaries, which were apparently blended
together at some point prior to the 11th century.
310
• Appendix D: The Lunyu entry in the “Yiwen zhi 藝文志” (“Monograph on Arts and Letters”)
(Hanshu 30.1716–1717)
• The ancient-script Lunyu in 21 sections. It came out of the walls of Kongzi’s [house]. It has two
“Zizhang” [chapters, i.e., book 19].
• The Qi version in 22 sections. It also has the sections “Asking about being a true king” and
“Understanding the Way.”
• The Lu version in 20 sections, with a commentary in 19 sections.
• Explanations of the Qi version in 29 sections.
• Xiahou [Sheng’s] explanations of the Lu version in 21 sections.
• The Marquis of Anchang’s [i.e., Zhang Yu’s 張禹] explanations of the Lu version in 21
sections.620
• Wang Jun’s explanations of the Lu version in 20 sections.
• The Yan commentary and explanations in 3 scrolls.
• Discussions and memorials in 18 sections.
• The Kongzi Family Sayings in 27 scrolls.
• Kongzi’s Three Audiences in 7 sections.
• The Charts and Models of Kongzi’s Followers in 2 scrolls.
—A total of 12 Lunyu textual lineages in 229 sections.
The Lunyu consists of Kongzi’s responses to his disciples and contemporaries as well as conversations between his disciples and the words they themselves heard from the Master. At that
time, [Kongzi's] followers each had his own record. After the Master died, his followers gathered
[his sayings] together and selectively edited [the collection], thus they called it the Selected Sayings. With the rise of the Han there were the explanations of the Qi and Lu versions. Those who
transmitted the Qi version were Commandant of the Capital of Changyi Wang Ji, Privy Treasurer
Song Ji, Grandee Secretary Gong Yu, Prefect of the Masters of Writing Wulu Chongzong, and
Master Yong of Jiaodong, but only Wang Ziyang [i.e. Wang Ji] was a famous expert. Those who
transmitted the Lu version were Gong Fen, Chief Commandant of Changshan, Xiahou Sheng,
620.The biography of Zhang Yu (Hanshu 81.3352) includes additional information about his Lunyu text:)
At first when Zhang Yu was serving as tutor he submitted a chapter-and-verse Lunyu commentary to the throne
because the Emperor [Cheng] had put several questions to him about the text. Initially, Fu Qing of Lu, Xiahou
Sheng, Wang Yang [=Wang Ji 王吉], Xiao Wangzhi, and Wei Xuancheng all explained the Lunyu, and the
arrangement of sections in their texts was somewhat different. Zhang Yu first served Wang Yang and afterwards
followed Master Sheng. He adopted their strong points and eventually put out [his own version] that was greatly
esteemed. The Ru said of it, “Want a Lunyu? Read Zhang Yu’s.” From this point forward scholars usually followed
Zhang’s text and the other lineages gradually declined.
初，禹為師，以上難數對己問經，為論語章句獻之。始魯扶卿及夏侯勝、王陽、蕭望之、韋玄成皆說論
語，篇第或異。禹先事王陽，後從庸生，采獲所安，最後出而尊貴。諸儒為之語曰：欲為論，念張文。由
是學者多從張氏，餘家寖微。
311
Privy Treasurer of Changxin, Chancellor Wei Xian, Fu Qing of Lu, General of the Van Xiao
Wangzhi, Zhang Yu, Marquis of Anchang. All of these men were famous experts. Mr. Zhang's
was the last [whose text] circulated at large.621
論語古二十一篇。出孔子壁中，兩子張。齊二十二篇。多問王、知道。魯二十篇，傳十九
篇。齊說二十九篇。魯夏侯說二十一篇。魯安昌侯說二十一篇。魯王駿說二十篇。燕傳說
三卷。議奏十八篇。石渠論。孔子家語二十七卷。孔子三朝七篇。孔子徒人圖法二卷。凡
論語十二家，二百二十九篇。
論語者，孔子應答弟子、時人及弟子相與言，而接聞於夫子之語也。當時弟子各有所
記。夫子既卒，門人相與輯而論篹，故謂之論語。漢興，有齊、魯之說。傳齊論者，昌邑
中尉王吉、少府宋畸、御史大夫貢禹、尚書令五鹿充宗、膠東庸生，唯王陽名家。傳魯論
語者，常山都尉龔奮、長信少府夏侯勝、丞相韋賢、魯扶卿、前將軍蕭望之、安昌侯張
禹，皆名家。張氏最後而行於世。
621.Titles from Bielenstein 1980.
312
• Appendix E: Lunyu transmitters mentioned in the “Yiwen zhi”
dates
associated texts
Zhang Yu 張禹
d. 5 BCE
Explanations (shuo 說) on the Lu Lunyu
Wang Jun 王駿
d. 15
Explanations (shuo 說) on the Lu Lunyu
Wei Xuancheng 韋玄成 d. 36
“explained (shuo 說) the Lunyu”622
Wulu Chongzong
fl. 38
Qi Lunyu
Gong Yu 貢禹
d. 44
Qi Lunyu; authored a “chapter-and-verse” commentary
Song Ji 宋畸
fl. 72–67
Qi Lunyu
Xiao Wangzhi 蕭望之
fl. 74–46
Lu Lunyu
Xiahou Sheng 夏侯勝
fl. 74–49
Explanations (shuo 說) on the Lu Lunyu
Yong Sheng 庸生
fl. 74–53
Qi Lunyu
Wang Ji 王吉
fl. 80–47
Qi Lunyu
Wei Xian 韋賢
d. 67
Lu Lunyu
五鹿充宗
Fu Qing of Lu 魯扶卿623 fl. c. 106
Lu Lunyu
Gong Fen 龔奮
Lu Lunyu
unknown
622.Wei Xuancheng is not mentioned in the “Yiwen zhi” but he is mentioned at Hanshu 81.3352 in the biography of
Zhang Yu.
623.Whereas Ban Gu refers to this figure ambiguously as “Lu Fuqing,” Wang Chong names him as “Fu Qing of
Lu.” The Jingdian shiwen (SBCK 1/31b) also quotes Zheng Xuan 鄭玄 to the effect that Fu Qing was also known as
Fu Xian 扶先 or Fu xiansheng 扶先生 (Master Fu). On this point see also Makeham 1996, p. 11, n. 42.
313
• Appendix F: Wang Chong’s history of the Lunyu (Lunheng jiaoshi 81.1135–1139)
All those who would explain the Lunyu merely know how to explain the characters and language
without knowing the original number of sections. They know that in the Zhou one foot was eight
inches without knowing why the [strips on which the] Lunyu [was written] were only one foot in
length. The Lunyu is the disciples’ collected records of Kongzi’s words and deeds. Because they
recorded his instructions so often there were several dozen to a hundred sections of text. They
used an eight-inch foot in order to economize the record-keeping and make it easier to carry. Because these remnants were not canonical they feared that the transmitted text and recorded
knowledge would be lost, so they only used an eight-inch foot [for the strips] and not the twofoot four-inch length.
At the rise of the Han the text was lost. During Emperor Wu’s reign an ancient-script version
in 21 sections was discovered in the walls of Kongzi’s [house]. The two versions of Qi and Lu
plus the nine sections of the Hejian text add up to 30 sections.624 Emperor Zhao read a text in 21
sections. Emperor Xuan sent it down to the Grand Master of Ceremonies and Erudites [because]
at that time the writing was still considered too difficult to understand. They [added explanations
and] called it a “commentary.”625 Afterwards it was rewritten in clerical script so that it could be
transmitted and recited. At first, Kongzi’s descendant Kong Anguo taught it to Fu Qing of Lu,
who was the first to call it the “Lunyu” after he was appointed Inspector of Jingzhou.
Today we speak of a Lunyu in 20 sections, but we have lost the Qi and Lu versions as well as
the Hejian version in 9 sections. Originally there were 30 sections, [some of which] were lost in
circulation. Some have [extant versions in] 21 sections, some have [versions with] more or fewer
section titles, and in their appraisal of the text some are correct and some are incorrect. Those
who would explain the Lunyu only know how to slice and dice with their questions and weave
trivialities with their contrived problems. They do not know to ask about the original number of
sections and section titles. “Keeping the old warm and knowing the new—this is how to be a
teacher.”626 How can one be called a teacher without knowing the ancient?
說論者，皆知說文解語而已，不知論語本幾何篇；但〔知〕周以八寸為尺，不知論語所獨
一尺之意。夫論語者，弟子共紀孔子之言行，敕記之時甚多，數十百篇，以八寸為尺，紀
之約省，懷持之便也。以其遺非經，傳文紀識恐忘，故但以八寸尺，不二尺四寸也。
漢興失亡。至武帝發取孔子壁中古文，得二十一篇，齊、魯二，河間九篇，三十篇。
624.There seem to be major problems with this section of the text, beginning with this line. My translation follows
the suggested emendations in the Lunheng jiaoshi.
625.The Hanshu biography of Xiahou Sheng 夏侯勝 records that Emperor Xuan asked him to produce a Lunyu
“explanation” (shuo 說) text. See Hanshu 75.3159.
626.This is an implicit quotation of Lunyu 2/11.
314
至昭帝女627讀二十一篇。宣帝下太常博士，時尚稱書難曉，名之曰傳；後更隸寫以傳誦。
初，孔子孫孔安國以教魯人扶卿，官至荊州刺史，始曰論語。
今時稱論語二十篇，又失齊、魯、河間九篇。本三十篇，分布亡失；或二十一篇。
〔篇〕目或多或少，文讚或是或誤。說論語者，但知以剝解之問，以織微之難，不知存問
本根篇數章目。溫故知新，可以為師；今不知古，稱師如何？
627.The word nü 女 is probably a mistake.
315
• Appendix G: A Shuiyuan 說苑 Kongzi attribution
Kongzi said, “He is acceptable for his simplicity.” “Simplicity” means unsophistication, and “unsophistication” means lacking ritual and patterning. When Kongzi went to see Zisang Bozi,
Zisang Bozi was there without a cap. One of Kongzi’s followers said, “Master, why would you
see this man?” Kongzi said, “His substance is fine though he is unrefined. I wish to persuade him
to refine himself.” After Kongzi had left, Zisang Bozi’s retainers were unhappy and said, “Why
would you see Kongzi?” He said, “His substance is fine though he is too refined. I wish to persuade him to unrefine himself.” Thus it is said, “One who cultivates his patterning and substance
is called a ‘noble man,’ and one who has substance but no patterning is called a ‘simple rustic.’”
Zisang Bozi was a simple rustic who wished to share the way of men with the oxen and horses.
That is why Zhonggong called him “greatly simplistic.” When above there is not an enlightened
son of heaven and below there are no worthy vassals then the world loses the Way. Subjects kill
their lords, sons kill their fathers, and if one has the power to attack another then the attack is acceptable. In Kongzi’s time there was not an enlightened son of heaven and so he said [in Lunyu
6/1], “Yong [=Zhonggong] could be made to face south.” “Facing south” means being son of
heaven. The reason why Yong was praised as one who could “face south” was because he asked
Kongzi about Zisang Bozi. Kongzi said, “He is acceptable for his simplicity.” Zhonggong said,
“Is it not acceptable to conduct oneself with respect but act simplistically? Is it not too simplistic
to conduct oneself simplistically yet also act simplistically?” The Master said, “Yong, your
words are right!” Zhonggong comprehended the techniques of transformation. Kongzi understood the royal Way and had nothing to add to Zhonggong’s words.
孔子曰：可也簡。簡者，易野也，易野者，無禮文也。孔子見子桑伯子，子桑伯子不衣冠
而處，弟子曰：夫子何為見此人乎？曰：其質美而無文，吾欲說而文之。孔子去，子桑伯
子門人不說，曰：何為見孔子乎？曰：其質美而文繁，吾欲說而去其文。故曰，文質脩者
謂之君子，有質而無文謂之易野，子桑伯子易野，欲同人道於牛馬，故仲弓曰太簡。上無
明天子，下無賢方伯，天下為無道，臣弒其君，子弒其父，力能討之，討之可也。當孔子
之時，上無明天子也，故言雍也可使南面，南面者天子也，雍之所以得稱南面者，問子桑
伯子於孔子，孔子曰：可也簡。仲弓曰：居敬而行簡以道民，不亦可乎？居簡而行簡，無
乃太簡乎？子曰：雍之言然！仲弓通於化術，孔子明於王道，而無以加仲弓之言。628
628.Shuiyuan 19 (SBCK 19/20a–b).
316
• Appendix H: Kongzi attributions in Western Han edicts
ruler
year
source quoted
quotation marker
Wang Mang 王莽
(r. 9 CE–23 CE)
20 CE
Lunyu 13/3
Xuan Ni gong yue 宣尼公曰 (“His
Resplendent Excellency Ni said...”)
(Hanshu 99c.4160)
zhuan bu yun hu 傳不云乎 (Does not
a tradition say...) (Hanshu 12.358)
2 CE
Lunyu 2/20, 3/4, Zhongni wei zhi 仲尼謂之 (“Kongzi
3/22, 12/17
said about him...”); Kongzi bu yun hu
孔子不云乎 (“Did not Kongzi
say...”) (Shiji 112.2963)
1 BCE Lunyu 13/2
none (Hanshu 12.348)
Grand Empress Dowager
Wang Zhengjun 王政君
(71 BCE–13 CE)
5 CE
Lunyu 8/2
Ai 哀 (r. 7–1 BCE)
5
7
Lunyu 3/14
Lunyu 15/11
none (Hanshu 11.339)
Kongzi bu yun hu 孔子不云乎 (“Did
not Kongzi say”) (Hanshu 22.1073)
Cheng 成 (r. 33–7)
16
23
Lunyu 15/30
Lunyu 15/10
28–25
Lunyu 6/10
fu 夫 (Hanshu 10.320)
none (but in the context of recruiting
official candidates) (Hanshu 10.313)
fuzi suo teng yue 夫子所痛，曰
(“this is what pained the Master when
he said”) (Hanshu 80.3319)
39
?
Lunyu 20/1629
Lunyu 15/30
Yuan 元 (r. 49–33)
zhuan yue 傳不云虖 (Hanshu 9.296)
Kongzi yue 孔子曰 (Kongzi said)
(Hanshu 80.3321)
Xuan 宣帝 (r. 74–49)
66
Lunyu 1/2
zhuan yue 傳曰 (a tradition says)
(Hanshu 8.250)
Emperor Wu 武帝
(r. 141–87 BCE)
123
Han Feizi (?)
Kongzi dui 孔子對 (Kongzi replied...)
(Hanshu 8.173)
128
Lunyu 5/28,
7/22 (?)
fu 夫 (Hanshu 6.166)
629.The same quotation also appears in the “Taishi 泰誓” chapter of the Shangshu (SBCK 6/4a).
317
• Appendix I: Emperor Wu’s 123 BCE Kongzi attribution and its parallels
Emperor Wu’s edict Lunyu 13/16
Han Feizi “Nan san 難三” (SBCK
(Hanshu 6.173)
16/2b)
Kongzi replied to
Duke Ding that
孔子對定公
that he should attract
the distant,
以徠遠，
The Duke of She asked about
governance.
Zigao, the Duke of She, asked
Zhongni about governance.
葉公問政。
葉公子高問政於仲尼，
The Master said, “Please the near Zhongni said, “Governance lies in
pleasing the near and attracting the
and attract the distant.”
distant.”
仲尼曰：政在悅近而來遠。
子曰：近者說，遠者來。
to Duke Ai that
哀公
Duke Ai asked Zhongni about
governance.
哀公問政於仲尼，
he should assess his
subjects,
以論臣，
Zhongni said, “Governance lies in
selecting the worthy.”
仲尼曰：政在選賢。
to Duke Jing that
景公
Duke Jing asked Zhongni about
governance.
齊景公問政於仲尼，
he should moderate
his expenses.
以節用，
Zhongni said, “Governance lies in
moderating expenses.
仲尼曰：政在節財。
It was not that their
desires were different,
it was that their problems demanded
different
solutions.
When the three lords had left
Zigong asked, “The three lords
asked you the same question about
governance, Master, yet your answers were different. Why is
that?”...
三公出，子貢問曰：
三公問夫子政一也，
夫子對之不同，何也？...
非期不同，所急異務
也。
318
319
◆
◆
◆ ◆
◆◆
Li Xun 李尋
0
★Yong
★谷永
◆
Wang Mang 王莽
◆◆
Gu
◆ Mei Fu 梅福
◆
◆
Chengdi
Zhang Song 張竦
Aidi
Pingdi
Zhai Fangjin 翟方進
◆◆
劉歆
Liu Xin
◆◆
Ping Dang 平當
175
200 CE
Fu Xie 傅夑
◆
anonymous
◆
★ Liu Xiang
◆ 劉向
Xun Shuang 荀爽
Ban Zhao 班昭
★ Xu Fang 徐防
Zuo Xiaong 左雄
Ma Rong 馬融
Ying Shao 應劭
Yuan Yan 爰延
Yuan Zhu 袁著
◆
Wang
Jia
王嘉
25
◆ ◆
★匡衡
Yuandi
◆
Gong Shengxing 公乘興
◆
◆
Du Qin
杜欽
Kuang
Heng
150
Jia Kui 賈逵
★ Yuan An 袁安
Xiandi
Lingdi
Huandi
Shundi
Andi
Hedi
Zhangdi
Mingdi
Guangwudi
Wang Mang
Zhang Min 張敏
◆
Wujiang Long 毋將隆
Yang Bing 楊秉
125
◆
★ Gong Yu 貢禹
◆
◆
Xuandi
Chen Chong 陳忠
Ban Biao 班彪
Du Lin 杜林
◆
50
★ Huo Guang 霍光
◆
Zhang Heng 張衡
Shang Min 尚敏
◆◆
Wei Xiang 魏相
◆
Zhang Fen 張奮 100
◆◆
◆
75
◆
◆
Yue Hui 樂
◆
◆
Zhaodi
Wudi
He Chang 何敞
75
50
25
◆
100
◆
Lu Pi 魯丕
Liu Cang 劉蒼
◆
125
Liang Tong 梁統
Chen
Yuan
陳元
◆
Yuqiu Shouwang 吾丘壽王
Jingdi
Wendi
Lu Kuang 魯匡 ★
◆ ◆
★ Dong Zhongshu 董仲舒
Jia Yi 賈誼
Gaozu
Huidi
◆
150
175
200 BCE
• Appendix J: Kongzi attributions in Han memorials (★=explicit Lunyu mention; ◆=use of a
Lunyu Kongzi saying)
◆
◆
◆
◆
• Appendix K: Lunyu Kongzi sayings in the Hanshu organized by Lunyu book (Numbers in boxes indicate the Hanshu chapter(s) which include(s) a particular Lunyu Kongzi saying)
Lunyu books 1–10
LY 1
/1
/2
/3
/4
/5
/6
/7
/8
/9
/10
/11
/12
/13
/14
/15
/16
/17
/18
/19
/20
/21
/22
/23
/24
/25
/26
/27
/28
/29
/30
/31
/32
/33
/34
/35
/36
/37
/38
/39
/40
/41
/42
/43
/44
21 23 56 90
LY 3
91
77 91
84
27
LY 4
LY 5
LY 6
LY 7
88
LY 8
28
92
24 86
LY 9
81 97
64
LY 10
88
25
41
28
23
24
30
56 88
60
56
77
73
81
99
80
83
75 81
22 88 99
99
99
22 56 88 100
21 81
28
22 88
30
99
64 88
36 86
64
20
20
25
22
22 56
56 64 99
27
46
56
28
36
66
75
29 67
20
60
20
56
LY 1
parallels
%
LY 2
3
19%
LY 2
3
13%
LY 3
11
42%
LY 4
4
15%
LY 5
10
36%
320
LY 6
6
20%
LY 7
LY 8
7
18%
5
24%
LY 9
8
26%
LY 10
1
4%
Lunyu books 11–20
LY 11
LY 12
LY 13
LY 14
LY 15
/1
LY 16
LY 17
LY 18
LY 19
24 74 83 86
21 30
/2
/3
56
23 30 83 99
87
36
/4
20
56
36 85 93
/5
/6
27 76
/7
30
30
56
6
21
22 27
99
99
56
85
/10
54
76
28 72
20
24
23 67
/11
23
22 88
/12
23 24 56 71
/13
30
72
/8
/9
LY 20
36
22
56
48 90
30
/14
/15
51
67 84
/16
/17
73
/18
/19
45 60
91
56
/20
23
99
54
/21
67
/22
36
/23
27
/24
/25
6 30 83
30
/26
56 99
30
/27
/28
/29
22 56
/30
23 30
80
/31
/32
/33
/34
/35
/36
/37
/38
/39
28
/40
/41
/42
/43
/44
LY 11
LY 12
LY 13
LY 14
LY 15
LY 16
LY 17
LY 18
LY 19
LY 20
parallels
3
7
9
3
10
8
6
2
5
3
%
12%
29%
30%
7%
24%
57%
23%
18%
20%
100%
321
• Appendix L: Lunyu Kongzi sayings in the Shiji organized by Lunyu book
(Numbers indicate the Hanshu chapter[s] in which a particular Lunyu Kongzi saying appears; outlined boxes indicate Lunyu Kongzi sayings with parallels in the “Kongzi shijia” [Shiji book 47] and “Zhongni dizi liezhuan” [Shiji
book 67])
Lunyu books 1–10
LY 1
/1
/2
/3
/4
/5
/6
/7
/8
/9
/10
/11
/12
/13
/14
/15
/16
/17
/18
/19
/20
/21
/22
/23
/24
/25
/26
/27
/28
/29
/30
/31
/32
/33
/34
/35
/36
/37
/38
/39
/40
/41
/42
/43
/44
LY 3
LY 4
47
122
28
67
67
104
47
23
28
LY 5
LY 6
67
67
67
67
67
67
67
67x2
67
67
67
47
67
67
67
67
67
67
67
LY 7
LY 8
31
47
47
LY 9
LY 10
27 47 49
74
47
47
47
47
47
47
47
47
47
47
67
47
47
74
47
47
47
61
67
47
47
67
47
47
47
47
47
47
47x2 121
61
112
47
47
47
47
47
47
103
47
67
61
47
67
47
LY 1
parallels
%
LY 2
2
13%
LY 2
5
21%
LY 3
10
38%
LY 4
2
8%
LY 5
14
50%
322
LY 6
10
33%
LY 7
16
42%
LY 8
1
5%
LY 9
9
29%
LY 10
7
26%
0
0%
Lunyu books 11–20
LY 11
/1
/2
/3
/4
/5
/6
/7
/8
/9
/10
/11
/12
/13
/14
/15
/16
/17
/18
/19
/20
/21
/22
/23
/24
/25
/26
/27
/28
/29
/30
/31
/32
/33
/34
/35
/36
/37
/38
/39
/40
/41
/42
/43
/44
parallels
%
67
LY 12
LY 13
LY 14
LY 15
67
67
67
67
67
67
47 74
47
47
23 47
67
67
67
LY 16
LY 17
LY 19
LY 20
67
47
67
67
47
67
47
67
LY 18
38
47
47
47
47
47 121
3
47
10
67
67 122
67
67
112
67
67
47
67 85
47
67
67
47
67
67
67
67
67
47
67
61
LY 11 LY 12 LY 13 LY 14 LY 15 LY 16 LY 17 LY 18 LY 19 LY 20
16
10
6
4
7
0
5
4
0
0
62%
42%
20%
9%
17%
0%
19%
36%
0%
0%
323
• Appendix M: Lunyu parallels in Huainanzi Kongzi sayings
Parallel 1
Huainanzi book 9 (SBCK 9/13b)
Lunyu 13/6
...When a ruler sets up laws he first makes
himself a model and exemplar, thus his orders
are enacted throughout the realm.
Kongzi said, “When one is correct in himself
[his will] is carried out without having to issue
any orders; when he is not correct in himself
[his will] is not followed even when he issues
orders.”
Thus when prohibitions prevail in himself his
orders are enacted among the people.
The Master said, “When one is correct in himself [his will] is carried out without having to
issue any orders; when he is not correct in himself [his will] is not followed even when he issues orders.”
...人主之立法，先自為檢式儀表，故令行於
天下。
孔子曰：其身正，不令而行；其身不正，雖 子曰：其身正，不令而行；其身不正，雖令
令不從。
不從。
故禁勝於身，則令行於民矣。
Parallel 2
Huainanzi book 13 (SBCK 13/11b)
Lunyu 9/30
Thus Kongzi said, “One can learn together
without being able to seek the Way [together];
one can seek the Way together without being
able to stand in it; one can stand in the Way
without being able to adapt it.” “Adapting” is
what the sages alone see.
The Master said, “One can learn together without being able to seek the Way [together]; one
can seek the Way together without being able
to stand in it; one can stand in the Way [together] without being able to adapt it.”
...故孔子曰：可以共學矣，而未可以適道
子曰：可與共學，未可與適道；可與適
也；可與適道，未可以立也；可以立，未可 道，未可與立；可與立，未可與權。
與權。權者，聖人之所獨見也。
324
Parallel 3
Lunyu 2/19632
Huainanzi book 16 (SBCK 16/11a)
When the Jisun clan seized control of the ducal
clan Kongzi tried to persuade them by first
conforming to their conduct and afterward entering their government.630
季孫氏劫公家，孔子說之。
先順其所為，而後與之入政。
Duke Ai asked, “What can I do to make the
people submit?”
哀公問曰：何為則民服？
Kongzi said,
曰：
“Raising the crooked in service of the
straight—
How can it not succeed?
Raising the straight in service of the
crooked—
That should not be followed.”
舉枉與直(*-əәk)，631
如何而不得(*-əәk)？
舉直與枉(*-aŋʔ)，
勿與遂往(*-aŋʔ)。
Kongzi responded,
孔子對曰：
“Raise the straight and set it against the
crooked
And the people will submit.
Raise the crooked and set it against the
straight,
And the people will not submit.
舉直錯諸枉(*-aŋʔ)，
則民服(*-əәk)；
舉枉錯諸直(*-əәk)，
則民不服(*-əәk)。
此所謂同汙而異途者。
This is what is called sharing the same defilement but for different ends.
630.In Sarah Queen’s and John Major’s translation of Huainanzi book 16 (2010, p. 650), the character 說 is read as
yue, i.e., “to be pleased.”
631.Here I read ju 舉 for yu 與 in keeping with the Huainanzi jishi edition (16.1140).
632.See also Lunyu 12/22 for the line “raising the straight against the crooked can make the crooked straight” (舉直
錯諸枉，能使枉者直。).
325
• Appendix N: Lunyu parallels in Lüshu chunqiu Kongzi sayings
Parallel 1
Lüshi chunqiu 4/2 (SBCK 4/5a)
Lunyu 11/23
When Kongzi was in distress in Kuang, Yan
Yuan followed after him.
Kongzi said, “I thought you had died.”
Yan Yuan said, “How could I dare die while
my master still lives?”
When the Master was in distress in Kuang,
Yan Yuan followed after him.
The Master said, “I thought you had died.”
Yan Yuan said, “How could I dare die while
my master still lives?”
孔子畏於匡，顏淵後，
孔子曰：吾以汝為死矣。
顏淵曰：子在，回何敢死？
子畏於匡，顏淵後。
子曰：吾以女為死矣。
曰：子在，回何敢死？
326
Parallel 2
Lüshi chunqiu 4/3 (SBCK 4/7a– Lunyu 7/2
Lunyu 7/34
b)
Zigong asked Kongzi, “How
will later ages praise you,
Master?”
The Master said, “
The Master said, “
Keep quiet and take note:
As for being sagely or humane,
how would I dare [aspire to
that]?
Kongzi said,
“How am I worthy of praise?
Enough of this talk.
learn without wearying of it
and teach others without tiring. It is rather that I conduct myFocus instead on loving
self without wearying and
learning without wearying of it
teach others without tiring.
and loving teaching without
What more is there for me?
tiring.
That is all that might be said of
me.
Let there only be this.”
Gongxi Hua said, “That is just
the reason why your disciples
cannot imitate you.”
子貢問孔子曰：後世將何以 子曰：
稱夫子？
默而識之，
學而不厭，誨人不倦，
孔子曰：
何有於我哉？
吾何足以稱哉？勿已者，
則好學而不厭，好教而不
倦，
其惟此邪。
327
子曰：
若聖與仁，則吾豈敢？
抑為之不厭，誨人不倦，
則可謂云爾已矣。
公西華曰：正唯弟子不能學
也。
Paragraph 3
Lüshi chunqiu 11/4 (SBCK 11/8a–b; KZJY 2.17/p.
Lunyu 13/18
20)
The Duke of She said to Kongzi,
“Among my people there is a certain Upright
In Chu there was a certain Upright Self who
Self who testified against his father after he
turned in his father after he had stolen a sheep. stole a sheep.”
The authorities arrested his father and were
about to execute him when Upright Self asked
to take his father’s place. When he was about
to be executed he said to the official, “I turned
my father in after he stole a sheep—am I not
trustworthy? I took my father’s place when he
was going to be executed—am I not filial? If
you would execute a trustworthy and filial
[man like me], who in this country would not
be executed?” The king of Chu heard about
this and did not execute Upright Self.
Kongzi said,
Kongzi heard about this and said,
“How strange is Upright Self’s trustworthiness! He uses his own father to make a good
name for himself.”
“The ‘upright’ among my people are different
from this. Fathers cover for their sons and sons
cover for their fathers. Uprightness lies in
this.”
Thus Upright Self’s trustworthiness is worse
than not being trustworthy at all.
葉公語孔子曰：
楚有直躬者，其父竊羊而謁之上，
吾黨有直躬者，其父攘羊，而子證之。
上執而將誅之。直躬者請代之。將誅矣，告
吏曰：父竊羊而謁之，不亦信乎？父誅而代
之，不亦孝乎？信且孝而誅之，國將有不誅
者乎？荊王聞之，乃不誅也。
孔子曰：吾黨之直者異於是：父為子隱，子
孔子聞之曰：異哉直躬之為信也，一父而載 為父隱。直在其中矣。
取名焉。
故直躬之信，不若無信。
328
Parallel 4
Lüshi chunqiu 14/6 (SBCK
Kongzi said, “...A noble man who achieves the
Way is said to have ‘achieved,’ and a noble
man who is in dire straits in [his pursuit of] the
Way is said to be ‘in dire straits’ Now I am
constrained by the Way of humaneness and
propriety, and this is why I face the calamities
of a chaotic age. How can this situation be
called “dire straits”? Thus reflecting within
myself I am not wanting in the Way; and in
facing such difficulties I have not lost my
virtue. When the extreme cold arrives and the
frost and snow fall I thereby know the flourishing of the pine and cypress. Previously Duke
Huan grasped this in Ju, Duke Wen in Cao, and
the king of Yue in Kuaiji. To me the distresses
of Chen and Cai are fortuitous!”
Lunyu 12/4
...The Master said, “If you examine yourself
within and are without remorse, then what is
there to worry about or fear?”
孔子曰：...君子達於道之謂達，窮於道之謂 ...子曰：內省不疚，夫何憂何懼？
窮。今丘也拘仁義之道，以遭亂世之患，其
所也，何窮之謂？故內省而不疚於道，臨難 Lunyu 9/28
而不失其德。
The Master said, “Only after the year grows
大寒既至，霜雪既降，吾是以知松柏之茂 cold does one know that the pine and cypress
也。
are the last to wither.”
昔桓公得之莒，文公得之曹，越王得之會
稽。陳、蔡之阨，於丘其幸乎！
子曰：歲寒，然後知松柏之後彫也。
329
• Appendix O: Lunyu parallels in Gongyang Kongzi sayings
Gongyang Ai 14 (SBCK 12/9a)
Lunyu 11/9
Yan Yuan died. The Master said, “Ah! Heaven Yan Yuan died. The Master said, “Ah! Heaven
has forsaken me!”
has forsaken me! Heaven has forsaken me!”
顏淵死。子曰：噫！天喪予！
顏淵死。子曰：噫！天喪予！天喪予！
330
• Appendix P: Lunyu parallels in Han Feizi Kongzi sayings
Han Feizi 38 (SBCK 16/2b; KZJY 10.41/pp. 226–27) Lunyu 13/16
When Zigao the Duke of She asked Zhongni
about governance Zhongni said, “Governance
lies in pleasing the near and attracting the
distant.”
When the Duke of She asked about governance
the Master said, “Please those near at hand and
attract those far away.”
When Duke Ai asked Zhongni about governance, Zhongni said, “Governance lies in selecting the worthy.”
When Duke Jing of Qi asked Zhongni about
governance, Zhongni said, “Governance lies in
moderating expenses.”
After the three dukes had departed Zigong
asked, “The three dukes asked you the same
question about government but your responses
were not the same. Why is that?”...
葉公子高問政於仲尼，
仲尼曰：政在悅近而來遠。
葉公問政。
子曰：近者說，遠者來。
哀公問政於仲尼，仲尼曰：政在選賢。
齊景公問政於仲尼，仲尼曰：政在節財。
三公出，子貢問曰：三公問夫子政一也，夫
子對之不同，何也？...
331
• Appendix Q: Lunyu parallels in Xunzi Kongzi sayings
Xunzi book 29 (SBCK 12a–13a; KZJY 3.16/p. 32)
Lunyu 2/17
Kongzi said,
The Master said,
“You [i.e., Zilu] take note of this! I say to you, “You [i.e., Zilu]! I will teach you about
making a show of your words is ostentatious- knowing.
ness, and making a show of your conduct is
braggadocio. The petty man tries to appear
wise and capable.
Thus when a noble man knows something he
says that he knows it, and when he does not
know something he says that he does not
know—this is the essence of speech.
Taking knowing as knowing and not knowing
as not knowing—this is knowing.”
When he can do something he says that he can
and when he cannot he says that he cannot—
this is the height of conduct. One who says
what is essential is wise, and one whose conduct is superior is humane. When one is both
humane and wise how could he have any
deficiency.
孔子曰：由志之！
子曰：由！誨女知之乎！
吾語汝：奮於言者華，奮於行者伐，色知而
有能者，小人也。
故君子知之曰知之，不知曰不知，言之要 知之為知之，不知為不知，是知也。
也；能之曰能之，不能曰不能，行之至也。
言要則知，行至則仁；既仁且知，夫惡有不
足矣哉！
332
Xunzi book 31 (SBCK 20/19a; KZJY 9.15/pp. 173–
Lunyu 15/6
74)
Zizhang asked about carrying out [one’s Way].
...Kongzi responded,
“A ‘noble man’ in his words is sincere and
trustworthy but in his heart does not feel his
own virtue. He is humane and righteous in his
person without having a haughty
appearance...”
...孔子對曰：所謂君子者，言忠信而心不
德。仁義在身而色不伐，思慮明通而辭不
爭，故猶然如將可及者，君子也。...
The Master said,
“If one’s words are sincere and trustworthy and
his conduct is earnest and respectful, then
[one’s Way] will be carried out even in barbarian lands...”
子張問行。子曰：言忠信，行篤敬，雖蠻貊
之邦，行矣。...
333
• Appendix R: Lunyu parallels in Zhuangzi Kongzi sayings
Parallel 1
Zhuangzi book 4 (SBCK 2/27b; KZJY 13.36/pp. 290– Lunyu 18/5
91)
When Kongzi was in Chu Jie Yu, the madman of Chu,
passed by his gates and said,
Jie Yu, the madman of Chu, sang as he passed by
Kongzi,
楚狂接輿歌而過孔子曰：
孔子適楚，楚狂接輿遊其門曰：
“Phoenix! Phoenix!
Why is your virtue diminished?
The past cannot be remonstrated with,
The future can still be chased after.
“Phoenix! Phoenix!
Why is your virtue so diminished?
The future cannot be awaited,
The past cannot be chased after.
When the world has the Way
The sage is realized;
when the world does not have the Way
the sage is born.
At the present time,
one can only avoid punishment.
Good fortune is lighter than a feather,
yet no one knows how to bear it.
Bad fortune is heaver than the earth,
Yet no one knows how to avoid it.
Enough! Enough!
You approach men with your virtue—
Perilous! Perilous!
You mark the ground as you hurry along—
You lose your yang! You lose your yang!
Do not hinder my progress!
My progress is winding.
Do not hinder my feet!
Enough! Enough!
Now those who follow good governance
are in peril.
鳳兮鳳兮！何德之衰(*-uj)？
往者不可諫(*-anʔ)，來者猶可追(*-uj)。
鳳兮鳳兮，何如德之衰(*-uj)也！
來世不可待(*-əәʔ)，往世不可追(*-uj)也。
天下有道(*-ûʔ)，聖人成(*-eŋ)焉；
天下無道(*-ûʔ)，聖人生(*-eŋ)焉。
方今之時(*-əәʔ)，僅免刑(*-eŋ)焉。
福輕乎羽(*-aʔ)，莫之知載(*-əәʔ)；
禍重乎地(*-ej)，莫之知避(*-ek)。
已乎已(*-əәʔ)乎！臨人以德(*-əәk)，
殆乎殆(*-əәʔ)乎！畫地而趨(*-o)！
迷陽迷陽(*-aŋ)，無傷吾行(*-aŋ)！
吾行郤曲(*-ok)，無傷吾足！(*-ok)
已而已(*-əәʔ)而！
今之從政者殆(*-əәʔ)而！
Kongzi descended and wished to speak with him. But
Jie Yu hurried off and avoided him and so would not
speak with him.
孔子下，欲與之言。趨而辟之，不得與之言。
334
Parallel 2
Zhuangzi 21 (SBCK 7/31a; KZJY 17.11/p. 390)
Lunyu 2/14
Yan Yuan asked Zhongni, “Master, when you
walk I walk, when you hurry I hurry, and when
you run I run. But when you leave me behind
in the dust I can only stare after you.”
The Master said, “Hui, what is this you’re
saying?”
The Master said,
He said, “Master, when you walk I walk—[that
means that] when you speak I speak. When
you hurry I hurry—[that means that] when you
debate I debate. When you run I run—[that
means that] when you speak of the Way I also
speak of the Way. As for leaving me behind in
the dust with me staring after you, [that means “The noble man is devoted without being overthat] you are trusted without saying anything, familiar, the petty man is overfamiliar without
you are devoted without being overfamiliar,
being devoted.”
and the people surge before you though you
have no instrument, but I simply do not know
how this is so.”
顏淵問於仲尼曰：夫子步亦步，夫子趨亦 子曰：
趨，夫子馳亦馳，夫子奔逸絕塵，而回瞠若
乎後矣！夫子曰：回，何謂邪？曰：夫子
步，亦步也；夫子言，亦言也；夫子趨，亦
趨也，夫子辯，亦辯也；夫子馳，亦馳也，
夫子言道，回亦言道也；及奔逸絕塵而回瞠
若乎後者，夫子不言而信，不比而周，無器 君子周而不比，小人比而不周。
而民滔乎前，而不知所以然而已矣。
Parallel 3
335
Zhuangzi 28 (SBCK 9/28a; KZJY 13.27/pp. 286–87)
Lunyu 12/4
Sima Niu asked about the noble man.
The Master said, “The noble man neither worries nor lives in fear.”
[Sima Niu] said, “Neither worrying nor living
in fear—is that all one needs to be called a noble man?”
司馬牛問君子。
子曰：君子不憂不懼。
曰：不憂不懼，斯謂之君子已乎？
...Kongzi said, “The noble man who passes
freely along the Way is called ‘passing freely.’
When he is exhausted in the Way he is called
‘exhausted.’ Now I embrace the Way of humaneness and propriety and encounter the
troubles of a chaotic world. How could this be
called ‘exhausted’? Thus by looking within
myself I am not exhausted in the Way, and
confronting difficulties I do not lose my virtue. The Master said, “If a man looks within himself and finds no faults, then what does he have
孔子曰：...君子通於道之謂通，窮於道之謂 to worry about or fear?”
窮。今丘抱仁義之道以遭亂世之患，其何窮
之為﹗故內省而不窮於道，臨難而不失其 子曰：內省不疚，夫何憂何懼？
德。
When the season has grown cold and the frost
and snow have fallen I know the flourishing of Lunyu 9/28
the pine and cypress. The straits of Chen and
Cai are a boon for me.”
The Master said, “Only after the year grows
cold does one know that the pine and cypress
are the last to wither.”
子曰：
歲寒，然後知松柏之後彫也。
天寒既至，霜雪既降，吾是以知松柏之茂
也。陳蔡之隘，於丘其幸乎。...
336
• Appendix S: Lunyu parallels in Mozi Kongzi sayings
Mozi 46 (SBCK 11/14b; KZJY 13.21/p. 281)
Lunyu 13/16
Zigao the Duke of She asked Zhongni about
governance, saying: “What sort of person is
good at governing?”
The Duke of She asked about governance.
葉公子高問政於仲尼曰：善為政者若之何？ 葉公問政。
Zhongni responded, “Someone who is good at The Master said, “Please those near at hand
governing attracts the distant and renews the
and attract those far away.”
old.”
子曰：近者說，遠者來。
仲尼對曰：善為政者，遠者近之，而舊者新
之。
Mozi learned of this and said, “Zigao did not
understand what to ask, and Zhongni did not
understand how to answer. How could Zigao
not know that those who are good at government attract the distant and renew the old? He
should have asked how this is accomplished.
[Kongzi] didn’t tell [Zigao] what he didn’t
know, he told him what he knew. That is why
Zigao did not understand what to ask, and
Kongzi did not understand how to answer.”
子墨子聞之曰：葉公子高未得其問也，仲尼
亦未得其所以對也。葉公子高豈不知善為政
者之遠者近也，而舊者新是哉？問所以為之
若之何也。不以人之所不智告人，以所智告
之，故葉公子高未得其問也，仲尼亦未得其
所以對也。...
337
Mozi book 39, “Fei ru xia 非儒下” (SBCK 9/22a;
Lunyu 10/12
KZJY 13.31/p. 289)
Duke Ai welcomed Kongzi, but if his seat was If his seat was not straight he did not sit.
not straight he would not sit, and if his food
was not cut properly he would not eat it.
席不正，不坐。
哀公迎孔丘，席不端弗坐，
割不正弗食。
Lunyu 10/8
...if his food was not cut properly he did not eat.
...割不正，不食。...
338
• Appendix T: Lunyu parallels in Zuozhuan Kongzi sayings
Parallel 1
Zuozhuan Zhao 12 (SBCK 22/15b; KZJY p. 418)
Lunyu 12/1
Yan Yuan asked about humaneness. The Master
said, “Overcoming oneself and returning to ritual is
humaneness. If in one day one can overcome himZhongni said, “There is an ancient maxim: ‘Over- self and return to ritual then the whole world will
coming oneself and returning to ritual is humane- return to humaneness. One becomes humane
ness.’ How true! If King Ling of Chu had been ca- through oneself—how could it be through someone
pable of this, how could he have been shamed at
else?
Ganxi?”
顏淵問仁。子曰：克己復禮為仁。一日克己復
仲尼曰：古也有志：克己復禮，仁也。信善
禮，天下歸仁焉。為仁由己，而由人乎哉？
哉。楚靈王若能如是，豈其辱於乾谿。
Yan Yuan said, “May I ask about the particulars?”
顏淵曰：請問其目。
The Master said, “If it is not ritually proper then do
not look at it; if it is not ritually proper then do not
hear it; if it is not ritually proper then do not say it;
if it is not ritually proper then do not act on it.”
子曰：非禮勿視，非禮勿聽，非禮勿言，非禮
勿動。
Yan Yuan said, “Even though I am not intelligent I
beg to put this saying into practice.”
顏淵曰：回雖不敏，請事斯語矣。
339
Parallel 2
Zuozhuan Ai 11 (SBCK 29/21a; KZJY p. 424)
Lunyu 15/1
Kongzi Wenzi was about to attack Dashu when he
called upon Zhongni.
Duke Ling of Wei asked Kongzi about military
formations.
Zhongni said,
Kongzi answered,
“I have learned about the hu and gui vessels but I
have never heard about military matters.”
“I have heard about the zu and dou vessels but I
have never learned about military affairs.”
Retiring he called for a carriage and departed.
The next day he departed.
孔文子之將攻大叔也。訪於仲尼。
衛靈公問陳於孔子。
仲尼曰：胡簋之事，則嘗學之矣。
甲兵之事，未之聞也。
孔子對曰：俎豆之事，則嘗聞之矣；
軍旅之事，未之學也。
退命駕而行。
明日遂行。
340
Parallel 3
Zuozhuan Ai 14 (SBCK 30/4a; KZJY p. 425)
Lunyu 14/21
On jiawu, Chen Huan of Qi assassinated his lord
Ren at Shuzhou. Kong Qiu purified himself for
three days and begged [Duke Ai] to attack Qi. After
the third entreaty the Duke said, “Lu is much weaker than Qi. If you would have us attack them then
would would happen?” He responded, “Chen Huan
assassinated his lord and half of the people do not
support him. We can defeat him with Lu’s strength
and half of Qi’s.
甲午，齊陳恆弒其君壬于舒州。孔丘三日齊而
請伐齊三，公曰：魯為齊弱久矣。子之伐之，
將若之何？對曰：陳恆弒其君，民之不與者
半。以魯之眾，加齊之半，可克也。
Chen Chengzi assassinated Duke Jian. Kongzi performed ablutions before going to court to inform
Duke Ai, “Chen Heng assassinated his lord. I beg
you to punish him.”
陳成子弒簡公。孔子沐浴而朝，告於哀公曰：
陳恆弒其君，請討之。
The Duke said, “Inform the Jisuns.”
The Duke said, “Tell the three sons [i.e., the heads
of the three great families].”
公曰：子告季孫。
公曰：告夫三子。
Kongzi declined then retired and announced to oth- Kongzi said, “I am ranked after the grandees and so
ers, “I am ranked after the grandees and so I did not I did not dare not to inform you. [Yet] My Lord
dare not to speak.”
said, ‘Tell the three sons.’ Kongzi reported this to
the three sons but his request was denied.
孔子辭，退而告人曰：吾以從大夫之後也。故
不敢不言。
孔子曰：以吾從大夫之後，不敢不告也。君曰
告夫三子者！之三子，告，不可。
Kongzi said, “I am ranked behind the grandees and
so I did not dare not to inform the duke.”
孔子曰：以吾從大夫之後，不敢不告也。633
633.Lunyu 11/8 also includes the line 以吾從大夫之後 but in a very different context.
341
• Appendix U: Lunyu 13/16 parallels in early texts
Mozi book 10
Mozi said, “...Pacify those near at hand and attract those far away.”
(SBCK 2/18a)
子墨子言曰：...近者安之，遠者歸之。
Mozi book 35
(SBCK 9/3b)
Mozi said, “...Pacify the governance of those near at hand and attract those far
away to your virtue.”
子墨子曰：...近者安其政，遠者歸其德。
Xunzi book 27 Zengzi said, “When those near at hand are pleased they feel close, when those
(SBCK 27/18a)
far away are pleased they attach to you; to make those near at hand feel close
and the distant attach is the way of the filial son.”
曾子曰：...近者說則親，遠者悅則附；親近而附遠，孝子之道也。
Guanzi book
An enlightened ruler causes the distant come to him and the near to feel close.
64 (SBCK 20/5b)
明主之使遠者來而近者親也。
Wenzi book 7
...please the near and attract the distant.
(Wenzi shuyi
7.337)
使近者悅，遠者來。
342
• Appendix V: Lunyu parallels in Shanghai Museum corpus Kongzi sayings
Parallel 1
Shanghai Museum/Guodian “Zi yi”, strip #13
Lunyu 2/3
(Shanghai bowuguan cang Zhanguo Chu zhushu vol. 1,
p. 188)
The Master said,
The Master said,
“When leading the people teach them with virtue
and keep them in line with ritual and they will
feel ? in their hearts. Teach them with governance
and keep them in line with punishments and the
people will be evasive in their hearts....”
“When you guide them with governance and keep
them in line with punishments the people will be
evasive and shameless. When you guide them with
virtue and keep them in line with ritual they will
feel shame and correct themselves.”634
子曰：
長民者教之以德，齊之以禮，
則民有？心。教之以政，齊之以刑，則民有免
心。
子曰：
道之以政，齊之以刑，民免而無恥；道之以
德，齊之以禮，有恥且格。
Parallel 2
Shanghai Museum/Guodian “Zi yi”, strip #23
Lunyu 13/22
(Shanghai bowuguan cang Zhanguo Chu zhushu vol. 1,
p. 198)
The Master said,
The Master said,
“The people of Song have a saying,
‘A man with no constancy cannot be a diviner.”
This saying survives from the ancients. If the tortoise-shells and milfoil stalks cannot be known,
how can people be known?...”
“Southerners have a saying,
‘A man with no constancy cannot be a diviner.’
A fine [saying] indeed!”
子曰：
子曰：
宋人有言曰：人而亡恒，〔不可為卜筮也，其 南人有言曰：人而無恆，不可以作巫醫。善
古之遺言與？龜筮猶弗知，而況於人乎？...〕635 夫！...
634.Translation adapted from Lau 2000, p. 11.
635.Brackets 〔〕 mark text missing in the Shanghai Museum version that has been reconstructed based on the
Guodian “Zi yi” manuscript.
343
Parallel 3
Shanghai Museum “Zhonggong 仲弓” (Shanghai Lunyu 13/2
bowuguan cang Zhanguo Chu zhushu vol. 3, pp. 263–
83)
When Ji Huanzi made Zhonggong chief steward,
Zhonggong informed Kongzi of [his appointment]...Zhonggong said, “May I ask what should
be prioritized in the practice of government?”
Zhongong was serving as chief steward for the Ji
clan when he asked Kongzi about governance.
仲弓為季氏宰，問政。
季桓子使仲弓為宰，仲弓以告孔子。...仲弓曰：
敢問為政何先？
The Master said, “Place office-holders first, forgive
Kongzi said, “...treat the elderly as elderly and be small offenses, and promote the talented and
kind to the young. Put office-holders first, promote worthy.”
the worthy and talented, and forgive trespasses and
crimes...”
孔子曰：...老老慈幼，先有司，舉賢才，宥過與 子曰：先有司，赦小過，舉賢才。
罪。
Zhonggong said, “As for treating the elderly as elderly and being kind to the young I have now heard
your instruction. How can I ‘put office-holders
first’?” Zhongni said, “The people are comfortable
with the old and they value [those who are]
promoted...”
...仲弓曰：若夫老老慈幼，既聞命矣。夫先有司
為之如何？仲尼曰：夫民安舊而重舉...
...there is completion. This is why office-holders
must be placed first.” Zhonggong said, “I am not
intelligent, for even when there are worthy and tal- Zhonggong said, “How can I recognize and proented men I do not know to promote them. May I mote the talented and worthy?”
ask how to ‘promote the talented’?”
...有成，是故有司不可不先也。
仲弓曰：雍也不敏，雖有賢才，弗知舉也。敢
問舉才如之何？
曰：焉知賢才而舉之？
Zhongni said, “The worthy and talented cannot be
hidden. Promote those whom you recognize. As for
those whom you do not recognize, would others
discard them?...”
仲尼：夫賢才不可弇也。舉爾所知，爾所不
知，人其舍之者。...
Kongzi said, “Promote those you recognize. Would
others discard [truly talented and worthy men]
whom you do not recognize?”
曰：舉爾所知；爾所不知，人其舍諸？
344
Parallels 4 & 5
“Junzi wei li”, strip #1 (Shanghai bowuguan cang
Lunyu 7/6
Zhanguo Chu zhushu vol. 5, p. 254)
The Master said, “Keep your mind on the Way, deYan Yuan was in attendance on the Master when
pend on virtue, rely on humaneness, and roam
the Master said, “Hui! The noble man in his perfor- within the arts.”
mance of ritual relies on humaneness.”
子曰：志於道，據於德，依於仁，遊於藝。
顏淵侍於夫子。
夫子曰：“回！君子為禮，以依於仁。
Lunyu 12/1
Yan Yuan arose and asked, “I am not intelligent
and I am unable to abide by this even a little.”
顏淵作而答曰：回不敏，弗能少居也。
The Master said, “Sit down and I will tell you.
Yan Yuan asked about humaneness.
顏淵問仁。
The Master said, “Restraining oneself and returning
to ritual is being humane. If one day someone
would restrain himself and return to ritual then the
entire world would return to humaneness. The path
to humaneness is through oneself—how could it be
through someone else?”
子曰：克己復禮為仁。一日克己復禮，天下歸
仁焉。為仁由己，而由人乎哉？
Yan Yuan said, “May I ask about the particulars?”
顏淵曰：請問其目。
The Master said, “If it is not ritually proper then do
If what you say will not be proper then your mouth not look at it; if it is not ritually proper then do not
will not say it; if what you see will not be proper
hear it; if it is not ritually proper then do not say it;
then your eyes will not see it; if what you hear will if it is not ritually proper then do not act on it.”
not be proper then your ears will not hear it.”
子曰：非禮勿視，非禮勿聽，非禮勿言，非禮
夫子曰：坐，吾語汝。言之而不義，口勿言
勿動。
也；視之而不義，目勿視也；聽之而不義，耳
勿聽也；動而不義，身毋動焉。
Yan Yuan said, “Even though I am not intelligent I
beg to put this saying into practice.”
顏淵曰：回雖不敏，請事斯語矣。
345
Parallel 6
“Dizi wen 弟子問”, strip #4 (Shanghai bowuguan
Lunyu 14/35
cang Zhanguo Chu zhushu vol. 5, p. 269)
The Master sighed and said, “Alas! There is no one The Master said, “There is no one who recognizes
who recognizes me!”
me!”
子嘆曰：“烏！莫我知也夫！
子曰：莫我知也夫！
Ziyou said, “Are you speaking about enacting [your Zigong said, “How is it that no one recognizes
Way]?”
you?”
子游曰：有施之謂也乎？
子貢曰：何為其莫知子也？
The Master said, “I do not resent heaven and I do
not fault others. In my studies, I start from below
and get through to what is up above. If I am understood at all, it is, perhaps, by heaven.”636
子曰：「不怨天，不尤人，下學而上達。知我
者其天乎！」
Parallel 7
“Dizi wen 弟子問” (Shanghai bowuguan cang
Lunyu 1/3
Zhanguo Chu zhushu vol. 5, p. 281)
[The Master?] said, “Crafty words and a ? appearance cannot be called ‘humane.’”
The Master said, “Crafty words and a contrived appearance are rarely humane.”
曰：巧言X（令？）色，未可謂仁也。
子曰：巧言令色，鮮矣仁！
636.Translation adapted from Lau 2000.
346
• Appendix W: Lunyu parallels in Mengzi Kongzi sayings
Parallel 1
Mengzi 2A/2 (SBCK 3/8b–9a; KZJY p. 510)
Lunyu 11/3
Lunyu 7/34
[Gongsun Chou] said,
“Zaiwo and Zigong were good at persuading and crafting phrases. Ran Niu, Minzi, and Yan Yuan were good
at speaking of virtuous conduct. Kongzi combined both
[skills], saying, ‘When it comes to rhetoric I have no
ability.’ That being so, then aren’t you, Master, already
a sage?”
Virtuous conduct: Yan Yuan,
Min Ziqian, Ran Boniu, and
Zhonggong. Speech: Zaiwo,
Zigong. Government affairs:
Ran You, Ji Lu. Refinement
and learning: Ziyou, Zixia.
[公孫丑]曰：宰我、子貢善為說辭；冉牛、閔子、顏 德行：顏淵、閔子騫、冉伯
淵善言德行。孔子兼之，曰：我於辭命，則不能
牛、仲弓。言語：宰我、子
也。然則夫子既聖矣乎？
貢。政事：冉有、季路。文
學：子游、子夏。
[Mengzi] said, “What is this you’re saying?!? Formerly
Zigong asked Kongzi, ‘Is the Master a sage?’
[孟子]曰：惡！是何言也？昔者子貢問於孔子曰：夫
子聖矣乎？
Lunyu 7/2
Kongzi said,
The Master said, “
The Master said, “
‘I am not capable of being a sage.
Be quiet and take note:
As for being sagely or
humane, how would I
dare [aspire to that]?
I learn without wearying and teach with tiring.’
learn without wearying of it
and teach others without
tiring.
It is rather that I conduct
myself without wearying
and teach others without
tiring.
What more is there for me?
That is all that might be
said of me.
子曰：
默而識之，
學而不厭，誨人不倦，
何有於我哉？
子曰：
若聖與仁，則吾豈敢？
抑為之不厭，誨人不
倦，則可謂云爾已矣。
孔子曰：
聖則吾不能，
我學不厭而教不倦也。
Zigong said, ‘Being insatiable in learning is wisdom and
being tireless in teaching is humaneness. Because he is
both humane and wise the Master is already a sage.’
Now if being a sage is something Kongzi could not
abide, then what is this you’re saying?”
Gongxi Hua said, “That
is just the reason why
your disciples cannot
imitate you.”
公西華曰：正唯弟子不
能學也。
子貢曰：學不厭，智也；教不倦，仁也。仁且智，
夫子既聖矣（乎）。夫聖、孔子不居，是何言也？
347
Parallel 2
Mengzi 2A/7 (SBCK 3/16b; KZJY p. 511)
Lunyu 4/1
Kongzi said,
“Living in a humane [environment] is finest. If
one chooses not to dwell in humaneness how
can he obtain wisdom?”
Now humaneness is the honored rank of heaven and the peaceful abode of men. To not be
humane when there is nothing holding you
back is not wise. To not be humane or wise, to
not possess ritual or propriety, is to be a slave.
The Master said,
孔子曰：里仁為美。擇不處仁，焉得智？
子曰：里仁為美。擇不處仁，焉得知？
“Living in a humane [environment] is finest. If
one chooses not to dwell in humaneness how
can he obtain wisdom.”
夫仁、天之尊爵也，人之安宅也。莫之禦而
不仁，是不智也。不仁、不智，無禮、無
義，人役也。
348
Parallel 3
Mengzi 3A/2 (SBCK 5/3b; KZJY p. 511)
Lunyu 14/40
Mengzi said, “...
Zizhang said, “The Shu say, ‘Gao Zong confined
himself to his mourning hut and did not speak for
three years.’ What does this mean?”
Kongzi said,
The Master said,
“Why must it [only] be Gao Zong? The ancients
were all like this. When a lord died the hundred of‘When the ruler passes away [his heir] listens to the fices all together heeded the prime minister for
prime minister, he sips porridge, and he is a deep
three years.”637
black. When he assumes the throne he cries, and
none of those in the hundred offices who serve in 子張曰：
government do not grieve.’
書云：
高宗諒陰，三年不言。何謂也？
孟子曰：...
孔子曰：君薨，聽於冢宰，歠粥，面深墨，即 子曰：何必高宗，古之人皆然。君薨，百官總
位而哭，百官有司莫敢不哀。
己以聽於冢宰三年。
Lunyu 12/19:
Ji Kangzi asked Kongzi about governance, saying,
“What if one killed those without the Way in order
to cleave to those with the Way?” Kongzi responded, “How can you use killing in your governance?
[That is because] he goes before them. What the su- If you desire what is good the people will be good.
perior likes the inferior will always like even more. The virtue of the noble man is like wind and that of
The virtue of the noble man is like wind and that of the petty man is like grass. Grass will always bend
the petty man is like grass. Grass will always bend in the wind.”
in the wind. This depends on the heir.”
季康子問政於孔子曰：如殺無道，以就有道，
先之也。上有好者，下必有甚焉者矣。君子之 何如？孔子對曰：子為政，焉用殺？子欲善而
德風也；小人之德草也。草上之風，必偃。 民善矣。君子之德風，小人之德草。草上之
風，必偃。
是在世子。
637.Translation adapted from Lau 2000, p. 145,
349
Parallel 4
Mengzi 3A/4 (SBCK 5/13a; KZJY
Lunyu 8/19
Lunyu 8/18
Kongzi said,
The Master said,
The Master said,
“How great was Yao’s conduct as
ruler! It was Heaven that was
great, and it was Yao who modeled himself on it. His magnificence was such that the people
could not put a name on it. How
lofty was Shun as a ruler! He had
the whole world but was not attached to it.”
“How great was Yao’s conduct as
ruler! How lofty! It was heaven
that was great and it was Yao
who modeled himself on it. His
“How lofty were Shun and Yu in
magnificence was such that the
possessing the world but not bepeople could not put a name on it. ing attached to it.
How lofty! How dazzling were
his successes, and his patterned
display.”
孔子曰：
子曰：
大哉堯之為君！
惟天為大，惟堯則之。蕩蕩
乎！
民無能名焉！君哉舜也！
巍巍乎！
大哉堯之為君也！巍巍乎！
唯天為大，唯堯則之。蕩蕩
乎！
民無能名焉。
巍巍乎！
巍巍乎！
p. 511)
子曰：
其有成功也，煥乎！其有文章
〔也〕！
有天下而不與焉。
350
舜、禹之有天下也而不與焉。
Parallel 5
Mengzi 4A/14 (SBCK 7/11a; KZJY p. 513)
Lunyu 11/17
Mengzi said, “While he was steward to the Ji clan The Ji clan was wealthier than the Duke of Zhou,
[Ran] Qiu was not able to reform his virtue, but in- and [Ran] Qiu on their behalf collected levies and
stead doubled the grain tax.
even increased them further.
Kongzi said,
The Master said,
‘Qiu is not my disciple. All of you can sound the
drums and attack him.’
“He is not my disciple. All of you can sound the
drums and attack him.”
Looking at it in this way, all rulers who do not enrich [their people] by practicing humane governance will by cast aside by Kongzi. How much
more so those who use force and warfare?”
孟子曰：
求也為季氏宰，無能改於其德，而賦粟倍他
日。
季氏富於周公，而求也為之聚斂而附益之。
孔子曰：
子曰：
求非我徒也，小子鳴鼓而攻之可也。
非吾徒也。小子鳴鼓而攻之可也。
由此觀之，君不行仁政而富之，皆棄於孔子者
也，況於為之強戰？
351
Parallel 6
Mengzi 4B/29 (SBCK 8/12b; KZJY p. 513)
Lunyu 6/11
Yu and Ji in an orderly age thrice passed by their
own doors without entering. Kongzi deemed them
worthies.
The Master said,
Yanzi [i.e., Yan Hui] in a disorderly age lived in a
squalid alley and lived on a bowl of rice and a
gourd of water. Other men could not have endured
such anxiety but Yanzi did not change his joyful
attitude.
“How worthy is [Yan] Hui! He lived on a bowl of
rice and a gourd of water and lived in a squalid alley. Other men could not have endured the anxiety
of his situation but [Yan] Hui did not change his
joyful attitude. How worthy is Hui!”
Kongzi deemed him a worthy.
禹、稷當平世，三過其門而不入，孔子賢之。
子曰：
顏子當亂世，居於陋巷，
賢哉回也！
一簞食，一瓢飲；人不堪其憂，
顏子不改其樂，孔子賢之。
一簞食，一瓢飲，在陋巷，人不堪其憂，
回也不改其樂。賢哉回也！
Parallel 7
Mengzi 5B/7 (SBCK 10/14b; KZJY p. 515)
Wan Zhang said, “When his lord summoned him
Kongzi would go without waiting for a carriage.
That being the case, was Kongzi in the wrong?”
[Mengzi] said, “Kongzi was serving as an official
with official responsibilities, and he was being
summoned in an official capacity.”
萬章曰：
孔子君命召，不俟駕而行；
然則孔子非與？曰：孔子當仕有官職，而以其
官召之也。
Lunyu 10/20
When the ruler summoned [he] went without waiting for a carriage.
君命召，不俟駕行矣。
352
Parallel 8
Mengzi 7B/37 (SBCK 14/16a–18b; KZJY pp. 516–17)
Lunyu 5/22
a) Wan Zhang asked, “Kongzi was in Chen when he said,
The Master was in Chen when he
said,
‘Let us return home. My disciples are unrestrained and wild. They “Go home! Go home! My disciples
are aggressive and do not forget their beginnings.’ Why did
are unrestrained and wild. While
Kongzi think about the unrestrained men of Lu when he was in
they have great accomplishments for
Chen?
all to see, they do not know how to
prune themselves.638
萬章問曰：
孔子在陳曰：盍歸乎來！吾黨之小子狂簡，進取，不忘其 子在陳，曰：歸與！歸與！吾
初。
黨之小子狂簡，
斐然成章，不知所以裁之。
孔子在陳，何思魯之狂士？
Lunyu 13/21
b) Mengzi said,
“When Kongzi could not associate with [men of] the middle way, The Master said, “When one cannot
he always [associated with someone who was] unrestrained or
associate with [men of] the middle,
timid. The unrestrained are aggressive while the timid have things one always [associates with somethey will not do. How could Kongzi not desire the middle way?
one who is] unrestrained or timid.
He could not always find [such men], thus he thought of the next The unrestrained are aggressive
best thing.”
while the timid have things they will
not do.”
孟子曰：
孔子不得中道而與之，
子曰：不得中行而與之，
必也狂狷乎！
必也狂狷乎！
狂者進取，狷者有所不為也。
狂者進取，狷者有所不為也。
孔子豈不欲中道哉？不可必得，故思其次也。
c) “May I ask what sort of person can be called ‘unrestrained’?”
敢問何如斯可謂狂矣？
d) [Mengzi] said, “Men like Qin Zhang, Zeng Xi, and Mu Pi are
what Kongzi would call ‘unrestrained.’”
曰：如琴張、曾騌、牧皮者，孔子之所謂狂矣。
e) “Why were they called ‘unrestrained’?”
何以謂之狂也？
638.Translation adapted from Lau 2000, p. 43.
353
f) [Mengzi] said, “Their aims were grandiose. They would say,
‘The ancients! The ancients!’ Yet when one considers their conduct they could not be contained. When he could not find an unrestrained man, Kongzi wished to find and associate with unsullied
men. Such men are timid, and they are third best.
曰：其志嘐嘐然，曰：古之人，古之人。夷考其行，而不掩
焉者也。狂者又不可得，欲得不屑不絜之士而與之，是獧
也，是又其次也。
——————————————————————————
g) Kongzi said, “Those who pass my gates without entering my
chamber are the upstanding villager types. Upstanding villager
types are the thieves of virtue.”
孔子曰：
過我門而不入我室，我不憾焉者，其惟鄉原乎！
鄉原、德之賊也。
h) [Wan Zhang] said, “What sort of person can be called an ‘upstanding villager type’?”
曰：何如斯可謂之鄉原矣？
i) [Mengzi] said, “Why must they be so grandiose? When they
speak they give no thought to their actions and when they act they
give no thought to their words. Thus they say, ‘The ancients! The
ancients!’ Why must they go it alone? Being born in this time they
[should] act in this age. This is fine so long as one is good. [But]
fawningly they curry favor in their time. These are the upstanding
villager types.”
曰：何以是嘐嘐也？言不顧行，行不顧言，則曰：古之人，
古之人。行何為踽踽涼涼？生斯世也，為斯世也，善斯可
矣。閹然媚於世也者，是鄉原也。
354
Lunyu 17/13
The Master said, “Upstanding villager types are the thieves of
virtue.”
子曰：
鄉原、德之賊也。
j) Wanzi said, “If an entire village praises an upstanding man
then everywhere he goes he is upstanding. Why would Kongzi
consider him a thief of virtue?”
萬子曰：一鄉皆稱原人焉，無所往而不為原人，孔子以為德
之賊，何哉？
k) [Mengzi] said, “If you want to condemn him you have nothing
to point to, and if you want to criticize him you have nothing to
criticize. He shares in the prevailing customs and accords with his
corrupt age. At home he seems loyal and trustworthy and in his
conduct he seems honest and pure of heart. The masses all delight
in him and so he considers himself right. But he cannot accompany you on the way of Yao and Shun. Thus [Kongzi] said, “The
thief of virtue.”
——————————————————————————
Kongzi said,
“[I] hate what resembles [the good] but isn’t.
I hate weeds because I fear they will be confused with sprouts.
I hate flattery because I fear it will be confused with propriety.
I hate skilled speakers because I fear they will be confused with
trustworthiness.
I hate the tones of Zheng because I fear they will be confused with
true music.
I hate purple because I fear it will be confused with vermillion.
I hate upstanding villager types because I fear they will be confused with the virtuous.
A noble man merely returns to what is standard. When the
standard is corrected the common people are motivated, and when
the common people are motivated there is no depravity.”639
曰：非之無舉也，刺之無刺也，同乎流俗，合乎汙世，居之
似忠信，行之似廉絜，眾皆悅之，自以為是，而不可與入堯
舜之道，故曰德之賊也。
孔子曰：
惡似而非者：惡莠，恐其亂苗也；惡佞，恐其亂義也；惡利
口，恐其亂信也，惡鄭聲，恐其亂樂也；惡紫，恐其亂朱
也；惡鄉原，恐其亂德也。
君子反經而已矣。經正，則庶民興；庶民興，斯無邪慝矣。
639.Translation adapted from Lau 1970, pp. 202–204.
355
Lunyu 17/18
The Master said,
“I hate that purple robs vermillion.
I hate that the tones of Zheng are
confused with proper music.
I hate that skilled speakers overturn
the royal family.”
子曰：
惡紫之奪朱也，惡鄭聲之亂雅
樂也，惡利口之覆邦家者。
• Appendix X: Lunyu parallels outside the corpus of Kongzi sayings*
*Dark highlights mark attributions different than those found in the Lunyu; light highlights mark
the parallels themselves.
Part 1: Summary of Lunyu entries with unattributed or differently attributed parallels
Lunyu 1
1/2, 1/3, 1/6, 1/8, 1/9, 1/10, 1/11, 1/12, 1/13, 1/14, 1/15, 1/16
Lunyu 2
2/1, 2/5, 2/7, 2/10, 2/13, 2/14, 2/17, 2/19, 2/23, 2/24
Lunyu 3
3/4, 3/24
Lunyu 4
4/3, 4/5, 4/18, 4/25
Lunyu 5
5/8, 5/12, 5/19, 5/25, 5/27, 5/28
Lunyu 6
6/3, 6/11, 6/15, 6/17, 6/18, 6/23, 6/30
Lunyu 7
7/1, 7/2, 7/6, 7/8, 7/9, 7/10, 7/19, 7/22, 7/28, 7/34
Lunyu 8
8/4, 8/5, 8/9, 8/13, 8/20, 8/21
Lunyu 9
9/4, 9/9, 9/13, 9/16, 9/17, 9/23, 9/25, 9/26, 9/28
Lunyu 10 10/1, 10/2, 10/3, 10/4, 10/5, 10/6, 10/8, 10/12, 10/18, 10/20, 10/25, 10/26
Lunyu 11 11/3, 11/7, 11/11, 11/16, 11/24
Lunyu 12 12/1, 12/2, 12/4, 12/5, 12/6, 12/11, 12/14, 12/16, 12/17, 12/19, 12/22
Lunyu 13 13/3, 13/4, 13/10, 13/15, 13/16, 13/18, 13/20, 13/25, 13/28, 13/29, 13/30
Lunyu 14 14/1, 14/10, 14/16, 14/24, 14/26, 14/27, 14/35, 14/38, 14/40, 14/42
Lunyu 15 15/1, 15/18, 15/21, 15/24, 15/27, 15/30, 15/40, 15/41
Lunyu 16 16/2, 16/6, 16/11, 16/14
Lunyu 17 17/1, 17/6, 17/10, 17/11, 17/17, 17/19, 17/22, 17/24
Lunyu 18 18/2, 18/6, 18/7, 18/8
Lunyu 19 19/1, 19/3, 19/8, 19/11, 19/14, 19/19, 19/21, 19/22
Lunyu 20 20/2, 20/3
356
Part 2: Lunyu parallels outside the corpus of Kongzi sayings
Lunyu book 1
Lunyu 1/2
• Guanzi 26, “Jie 戒”
管仲對曰：...孝弟者，仁之祖也。
(SBCK 10/2a)
有子曰：
...君子務本，本立而
道生。孝弟也者，其
為仁之本與！
• Lüshi chunqiu 14.1,
“Xiao xing 孝行”
<unattributed>務本莫貴於孝。
(SBCK 14/1a)
• Shuiyuan 3, “Jian ben 孔子曰：君子務本，本立而道生。
建本” (SBCK 3/1a)
• Yantie lun 12, “You
<unattributed>夫欲安民富國之道，
bian 憂邊” (SBCK 2/17b– 在於反本，本立而道生。
18a)
Lunyu 1/3
子曰：
巧言令色，
鮮矣仁！
• Shangshu, “Jiong
王若曰：...無以巧言令色
ming 冏命” (SBCK 12/6a)
• Shangshu, “Gao Yao 皋陶曰:...何畏乎巧言令色孔壬。
mo 皋陶謨” (SBCK 2/7a)
• Yi Zhoushu 68, “Wu ji <unattributed>幣帛之閒，有巧言令色，
武紀” (Yi Zhoushu buzhu 事不成。車甲之閒，
zhuyi 68.425)
有巧言令色，事不捷。
• Yi Zhoushu 58, “Guan 周公曰：...華廢而誣，巧言令色，
ren 官人” (Yi Zhoushu
皆以無為有者也。此之謂考志。
buzhu zhuyi 58.331)
• Guanzi 45, “Ren fa 任
法” (SBCK 15/7a)
• Da Dai Liji 49,
“Zengzi li shi 曾子立事
” (SBCK 4/4a)
<unattributed>美者以巧言令色請其主，
主因離法而聽之，此所謂美而淫之也。
曾子曰：...巧言令色，能小行而篤，難於
仁矣。
• Da Dai Liji 72, “Guan 王曰：...華如誣，巧言、令色、足恭一
ren 官人” (SBCK 10/3b) 也，皆以無為有者也。此之為考志。
• Shuiyuan 2, “Chen shu <unattributed>三曰中實頗險，
臣術” (SBCK 2/2b)
外容貌小謹，巧言令色，又心嫉賢...
357
Lunyu 1/6
• Mengzi 3B/4 (SBCK
6/6a)
子曰：弟子入則孝，
出則悌，謹而信，汎
• Mozi 35, “Fei ming
愛眾，而親人。行有
shang 非命上” (SBCK
餘力，則以學文。
9/4a)
• Xunzi 29, “Zi dao 子
道” (SBCK 20/9a)
• Huainanzi 20 (SBCK
20/10a)
Lunyu 1/8
Lunyu 1/9
孟子曰：...於此有人焉，入則孝，出則
悌，守先王之道，以待後之學者，而不得
食於子。
<unattributed>是以入則孝慈於親戚，
出則弟長於鄉里。
入孝出弟，人之小行也。
<unattributed>孔子弟子七十，養徒三千
人，皆入孝出悌，言為文章，行為儀表，
教之所成也。
<see 9/25 & 15/30>
• Mao Shi, preface to
#165, “Fa mu 伐木”
<unattributed>親親以睦，友賢不棄，
不遺故舊，則民德歸厚矣。
曾子曰：慎終追遠， (SBCK 9/5b)
民德歸厚矣。
Lunyu 1/10
• Hanshi waizhuan 10
(SBCK 10/4a)
子禽問於子貢曰：
• Shiji 104.2770
夫子至於是邦也，
必聞其政，求之與？
抑與之與？...
Lunyu 1/11
子曰：...三年無改於
父之道，可謂孝矣。
Lunyu 1/12
<unattributed>君子溫儉以求於仁，
恭讓以求於禮，得之自是，不得自是。
太史公曰：孔子稱曰：
居是國必聞其政，田叔之謂乎！
• Da Dai Liji 50,
曾子曰：...父死三年，不敢改父之道。
“Zengzi ben xiao 曾子
本孝” (SBCK 4/8a)
• Da Dai Liji 70, “Yu
Dai de 虞戴德” (SBCK
公曰：先聖之道，斯為美乎？
子曰：斯為美。...
有子曰：禮之用，和 9/10b)
為貴。先王之道，斯
為美；小大由之。有
所不行，知和而和，
不以禮節之，亦不可
行也。
Lunyu 1/13
• Liji 32, “Biao ji 表記” 子曰：君子慎以辟禍，篤以不揜，
(SBCK 17/1b)
恭以遠恥。
有子曰：信近於義， • Liji 32, “Biao ji 表記” 子曰：恭近禮，儉近仁，信近情，
言可復也。恭近於
(SBCK 17/3b)
敬讓以行。
禮，遠恥辱也。...
• Shuiyuan 19 (SBCK
孔子曰：恭近於禮，遠恥辱也。
19/4a)
358
Lunyu 1/14
子曰：
君子食無求飽，
居無求安，
敏於事而慎於言...
• Zuozhuan Xi 4 (SBCK 大子[奚齊]曰：君非姬氏，
5/5a)
居不安，食不飽，我辭。
• Mao Shi 135,
<unattributed>於我乎，每食四簋，今也
“Quanyu 權輿” (SBCK 每食不飽，于嗟乎，不承權輿。
6/16a)
• Hanshi waizhuan 2
• Xinyu 10, “Ben xing
本行” (SBCK 下/9b)
子路曰：...曾子褐衣縕緒，未嘗完也，糲
米之食，未嘗飽也。
<unattributed>君子篤於義而薄於利，
敏於行而慎於言。
• Mozi 18, “Fei gong
zhong 非公中” (SBCK
是故子墨子曰：古者有語：謀而不得，則
以往知來，以見知隱。
(SBCK 2/15a)
Lunyu 1/15
...子曰：賜也，始可 5/2b–3a)
與言詩已矣，告諸往
而知來者。
Lunyu 1/16
• Guanzi 32,
“Xiao cheng 小稱”
管子曰：身不善之患，毋患人莫己。
子曰：不患人之不己 (SBCK 11/7a)
知，患不知人也。
Lunyu book 2
Lunyu 2/1
• Shiji 42.1774,
子產謂韓宣子曰：
子曰：為政以德，譬 “Zheng shijia 鄭世家” 為政必以德，毋忘所以立。
如北辰，居其所而眾
星共之。
Lunyu 2/5
...子曰：生，事之以 • Mengzi 3A/2 (SBCK
禮；死，葬之以禮， 5/2b)
祭之以禮。
Lunyu 2/7
...子曰：
今之孝者，
是謂能養。
至於犬馬，
皆能有養；不敬，
何以別乎？
曾子曰：生、事之以禮，死、葬之以禮，
祭之以禮，可謂孝矣。
• Lüshi chunqiu 14/1,
<unattributed>民之本教曰孝，其行孝曰
“Xiao xing 孝行” (SBCK 養。養可能也，敬為難。敬可能也，安為
14/3a–3b)
難。安可能也，卒為難。父母既沒，敬行
• Liji, “Ji yi 祭義”
其身，無遺父母惡名，可謂能終矣。
(SBCK 14/12a)
• Da Dai Liji, “Zengzi
ben xiao 曾子本孝”
(SBCK 4/10a)
• Liji, “Nei ze 內則”
曾子曰：...父母之所愛亦愛之，父母之所
敬亦敬之。至於犬馬盡然，而況於人乎。
(SBCK 8/21a)
359
Lunyu 2/10
• Mengzi 4A/15 (SBCK
孟子曰：
聽其言也，觀其眸子，人焉廋哉？
子曰：視其所以，觀 • Da Dai Liji 72,
文王曰：...聽其聲，處其氣，考其所為，
其所由，察其所安。 “Wen wang guan ren 文 觀其所由，察其所安；以其前占其後，
人焉廋哉？人焉廋
王官人” (SBCK 10/4a)
以其見占其隱，以其小占其大。
哉？
此之謂『視中』也。
7/11b–12a)
• Yi Zhoushu 58,
“Guan ren 官人” (Yi
周公曰：...考其所為，觀其所由，以其
前，觀其後，以其顯，觀其隱，
以其小，占其大，此之謂視聲。
Zhoushu jiaobu zhuyi
58.333)
Lunyu 2/13
• Da Dai Liji 49,
“Zengzi lishi 曾子立事
” (SBCK 4/1b)
子貢問君子。
子曰：先行其言而後 • Da Dai Liji 54,
從之。
“Zengzi zhi yan 曾子制
言” (SBCK 5/3a)
曾子曰：君子博學而孱守之，微言而篤行
之，行必先人，言必後人，
君子終身守此悒悒。
曾子曰：...且夫君子執仁立志，先行後
言，千里之外，皆為兄弟...
Lunyu 2/14
• Zhuangzi 21, “Tian
子曰：君子周而不
Zifang 田子方” (SBCK
比，小人比而不周。 7/31a)
[顏淵曰：] 夫子不言而信，不比而周，無
器而民滔乎前，而不知所以然而已矣。
Lunyu 2/17
子曰：由！誨女知之 • Xunzi 8, “Ru xiao 儒
乎！知之為知之，不 孝” (SBCK 4/17b)
知為不知，是知也。
Lunyu 2/19
哀公問曰：
何為則民服？孔子對
曰：舉直錯諸枉，則
民服；舉枉錯諸直，
則民不服。
<unattributed>知之曰知之，
不知曰不知，內不自以誣，外不自以欺，
• Wenzi 4, “Fu yan 符言 老子〔文子〕曰：舉枉與直，如何不得；
” (Wenzi yishu 4.213)
舉直與枉，勿與遂往；
所謂同污而異泥者。
Lunyu 2/23
• Shiji 99.2722,
叔孫通曰：五帝異樂，三王不同禮。禮
“Shusun Tong liezhuan 者，因時世人情為之節文者也。
子張問：十世可知
叔孫通列傳”
故夏、殷、周之禮所因損益可知者，謂不
也？子曰：殷因於夏
相復也。臣願頗采古禮與秦儀雜就之。
禮，所損益，可知
也...
360
Lunyu 2/24
• Liji, “Qu li xia 曲禮下 <unattributed>非有所祭而祭之，名曰淫
” (SBCK 1/24b)
祀，淫祀無福。
子曰：非其鬼而祭
之，諂也。...
Lunyu book 3
Lunyu 3/4
• Da Dai Liji 49,
“Zengzi li shi” (SBCK
4/3b)
林放問禮之本。
子曰：大哉問！禮，
與其奢也，寧儉；
喪，與其易也，寧
戚。
Lunyu 3/24
儀封人請見...出曰：
二三子何患於喪乎？
天下之無道也久矣...
• Shiji 47.1944,
“Kongzi shijia 孔子世
家”
曾子曰：...君子入人之國，不稱其諱，不
犯其禁，不服華色之服，不稱懼惕之言。
故曰：與其奢也寧儉，與其倨也寧句。
孔子...謂子貢曰：
天下無道久矣，莫能宗予。
Lunyu book 4
Lunyu 4/1
• Xunzi 27, “Da lüe 大
略” (SBCK 27/5b–6a)
<unattributed>義有門；仁、
非其里而處之，非仁也；
義，非其門而由之，非義也。
...君子處仁以義，然後仁也；行義以禮，
然後義也；制禮反本成末，然後禮也。
Lunyu 4/3
• Guoyu 18, “Chu yu
xia 楚語下” (SBCK
子高曰：不然。吾聞之，唯仁者可好也，
可惡也，可高也，可下也。
子曰：唯仁者能好
人，能惡人。
18/13a)
Lunyu 4/5
• Yanzi chunqiu (SBCK
子曰：里仁為美。
擇不處仁，焉得知？
6/14a)
子尾曰：
富者，人之所欲也，何獨弗欲？
子曰：富與貴，是人
之所欲也；不以其道
得之，不處也。...
Lunyu 4/18
• Da Dai Liji 53,
單居離問於曾子曰：事父母有道乎？曾子
“Zengzi shi fumu 曾子 曰：有。愛而敬。父母之行若中道，則
子曰：事父母幾諫， 事父母” (SBCK 4/12a)
從；若不中道，則諫；諫而不用，行之如
見志不從，又敬不
由己。從而不諫，非孝也；諫而不從，亦
違，勞而不怨。
非孝也。孝子之諫，達善而不敢爭辨；爭
辨者，作亂之所由興也。
361
Lunyu 4/25
• Hanshi waizhuan 10
(SBCK 10/1a)
子曰：德不孤，
必有鄰。
• Xinxu 4, “Za shi 雜事
” (SBCK 4/9a)
桓公曰：善哉！祝乎！寡人聞之矣：
至德不孤，善言必再。叟盍優之？
Lunyu book 5
Lunyu 5/8
孟武伯問：子路仁
• Shiji 67.2190,
季康子問孔子曰：冉求仁乎？
乎？子曰：不知也。 “Zhongni dizi liezhuan ...復問：子路仁乎？
又問。子曰：由也， 仲尼弟子列傳”
千乘之國，可使治其
賦也，不知其仁也。
Lunyu 5/12
• Guanzi 51, “Xiao wen 管子對曰：
小問” (SBCK 16/7b)
...非其所欲，勿施於人，仁也。
子貢曰：我不欲人之 • Mengzi 4A/9 (SBCK
孟子曰：...所欲與之聚之，所惡勿施，爾
加諸我也，吾亦欲無 7/8a)
也。
加諸人。子曰：賜
• Liji, “Zhong yong 中 子曰：...施諸己而不願，亦勿施於人。
也，非爾所及也。
庸” (SBCK 16/3b)
<see also Lunyu 12/2
& 15/24>
Lunyu 5/19
• Zhuangzi 21, “Tian
Zifang 田子方” (SBCK
子張問曰：令尹子文
三仕為令尹，無喜
色；三已之，無慍
色。舊令尹之政，必
以告新令尹。何如？
7/39b)
子曰：甯武子，
邦有道，則知；
邦無道，則愚。
其知可及也，
其愚不可及也。
• Mao Shi, commentary <unattributed>國有道則知，
to # 256, “Yi 抑” (SBCK 國無道則愚。
• Huainanzi 12, “Dao
ying 道應” (SBCK
肩吾問於孫叔敖曰：
子三為令尹而不榮華，三去之而無憂色。
<unattributed>昔孫叔敖三得令尹，
無喜志；三去令尹，無憂色。
12/17b)
• Huainanzi 13, “Fan
<unattributed>叔孫敖三去令尹
lun 氾論” (SBCK 13/19a) 而無憂色，爵祿不能累也
18/3b)
Lunyu 5/25
• Da Dai Liji 72, “Wen 王曰：華如誣，巧言、令色、足恭一也，
wang guan ren 文王官 皆以無為有者也。此之為考志。
子曰：巧言、令色、 人” (SBCK 10/3b)
足恭，左丘明恥之，
丘亦恥之。...
362
Lunyu 5/27
• Hanshi waizhuan 3
晉文公：...今不內自訟過，不悅百姓，將
何錫之哉！
(SBCK 3/17a)
子曰：已矣乎！吾未
見能見其過而內自訟
者也。
Lunyu 5/28
• Da Dai Liji 54,
曾子曰：...昔者禹見耕者五耦而式，
子曰：十室之邑，
“Zengzi zhi yan 曾子制 過十室之邑則下，為秉德之士存焉。
必有忠信如丘者焉， 言” (SBCK 5/5b–6a)
不如丘之好學也。
Lunyu book 6
Lunyu 6/3
• Da Dai Liji, “Wei
子貢對曰：夙興夜寐，諷誦崇禮；行不貳
jiangjun Wenzi 衛將軍 過，稱言不苟，是顏淵之行也。...不遷
哀公問：弟子孰為好 文子” (SBCK 6/4a)
怒，不深怨，不錄舊罪，是冉雍之行也。
學？
• Kongzi jiayu 12, “Dizi
孔子對曰：有顏回者 xing 弟子行” (SBCK
好學，不遷怒，不貳 3/5b)
過。不幸短命死矣， • Lunyu 11/7
季康子問弟子孰為好學。
今也則亡，未聞好學
孔子對曰：有顏回者好學，
者也。
不幸短命死矣，今也則亡。
Lunyu 6/11
• Mengzi 4B/29 (SBCK
<unattributed>顏子當亂世，居於陋巷，
一簞食，一瓢飲；人不堪其憂，
顏子不改其樂，孔子賢之。
• Xinyu 6, “Shen wei 慎 <unattributed>顏回一簞食，一瓢飲，
微” (SBCK 上/14a)
在陋巷之中，人不堪其憂，
回也不改其樂。
8/12b)
子曰：賢哉回也！一
簞食，一瓢飲，在陋
巷，人不堪其憂，回
也不改其樂。賢哉回
也！
Lunyu 6/15
• Zuozhuan Ai 11 (SBCK <unattributed>孟之側後入，以為殿。抽
29/18b)
矢策其馬曰：馬不進也。
子曰：孟之反不伐，
奔而殿，將入門，策
其馬，曰：非敢後
也，馬不進也。
Lunyu 6/17
• Liji, “Li qi 禮器”
<unattributed>未有入室而不由戶者。
(SBCK 7/15b)
子曰：誰能出不由
戶？何莫由斯道也？
363
Lunyu 6/18
• Xinshu, “Bing ju zhi
rong 兵車之容” (SBCK
子曰：質勝文則野， 6/9b)
文勝質則史。文質彬
彬，然後君子。
Lunyu 6/23
• Hanshi waizhuan 3
(SBCK 3/16a)
子曰：知者樂水，仁 • Kongzi jiayu 7, “Wu
者樂山。知者動，仁 yi 五儀” (SBCK 1/28b)
者靜。知者樂，仁者
壽。
Lunyu 6/30
• Hanshi waizhuan 7
<unattributed>語曰：審乎明王，執中履
衡。言秉中適而據乎宜。故威勝德則淳，
德勝威則施。威之與德，交若繆纆。且畏
且懷，君道正矣。質勝文則野，文勝質則
史，文質彬彬，然後君子。
<unattributed>問者曰：夫智者何以樂於
水也？...問者曰：夫仁者何以樂於山也？
哀公問於孔子曰：智者壽乎？仁者壽乎？
孔子對曰：然...
狐丘丈人曰：善哉！言乎！堯舜其猶病
諸！
[魏]文侯曰：善哉言乎！堯舜其猶病諸！
寡人雖不敏，請守斯語矣。
(SBCK 7/9a)
...子曰：何事於仁！ • Hanshi waizhuan 8
必也聖乎！堯、舜其 (SBCK 8/19a)
猶病諸！
Lunyu book 7
Lunyu 7/1
Lunyu 7/2
<see Appendix 3:P>
• Wenzi 10, “Shang yi
上仁” (Wenzi shuyi
老子〔文子〕曰：學而不厭，所以治身；
教而不倦，所以治民。
子曰：默而識之，學 10.457–58)
而不厭，誨人不倦，
何有於我哉？
Lunyu 7/3
子曰：德之不脩，學
之不講，聞義不能
徙，不善不能改，是
吾憂也。
Lunyu 7/6
<for 不善不能改 parallels, see 15/30>
• Guodian “Yu cong san <unattributed>志於道，狎於德，依於
語叢三” (Guodian Chu mu 仁，遊於藝。
子曰：志於道，據於 zhujian p. 211)
德，依於仁，遊於
• Liji, “Shao yi 少儀”
藝。
(SBCK 10/13b)
Lunyu 7/8
• Shiji 47.1938,
“Kongzi shijia 孔子世
子曰：不憤不啟，不 家”
悱不發。舉一隅不以
三隅反，則不復也。
364
<unattributed>士依於德，游於藝。
<unattributed>不憤不啟，舉一隅不以三
隅反，則弗復也。
Lunyu 7/9
• Liji, “Tan gong shang <unattributed>食於有喪者之側，
檀弓上” (SBCK 2/14b)
未嘗飽也。
子食於有喪者之側，
未嘗飽也。
Lunyu 7/10
• Liji, “Qu li shang 曲
禮上” (SBCK 1/14a)
<unattributed>里有殯，不巷歌，適墓不
歌，哭日不歌。
• Hanshi waizhuan 2
子夏讀詩已畢。夫子問曰：爾亦何大於詩
矣？子夏對曰：...亦可發憤忘食矣。
晏子曰：...且人何憂，靜處遠慮，見歲若
月，學問不厭，不知老之將至，
安用從酒！
子於是日哭，則不
歌。
Lunyu 7/19
(SBCK 2/17b)
...子曰：女奚不曰， • Yanzi chunqiu (SBCK
其為人也，發憤忘
6/13a)
食，樂以忘憂，不知
老之將至云爾。
Lunyu 7/22
• Zuozhuan Xiang 31
(SBCK 19/15b)
子曰：三人行，必有
我師焉，擇其善者而
從之，其不善者而改
之。
Lunyu 7/28
子產曰：何為？夫人朝夕退而游焉，以議
執政之善否。其所善者，吾則行之；其所
惡者，吾則改之。是吾師也。
• Shuiyuan 3, “Jian ben 吳子曰：...多聞而擇焉。
建本” (SBCK 3/1b)
子曰：...多聞，擇其
善者而從之；多見而
識之；知之次也。
Lunyu 7/34
• Mengzi 2A/2 (SBCK
昔者子貢問於孔子曰：夫子聖矣乎？孔子
曰：聖則吾不能，我學不厭而教不倦也。
3/9a)
子曰：若聖與仁，則
吾豈敢？抑為之不
厭，誨人不倦，則可
謂云爾已矣。...
Lunyu book 8
Lunyu 8/4
• Yantie lun 31, “Zhen
shi 箴石” (SBCK 6/11b)
曾子有疾，孟敬子問
之。曾子言曰：...君
子所貴乎道者三：動
容貌，斯遠暴慢矣；
正顏色，斯近信矣；
出辭氣，斯遠鄙倍
矣。...
365
丞相曰：吾聞諸鄭長【孫】【者】曰：
君子正顏色，則遠暴嫚；
出辭氣，則遠鄙倍矣。
Lunyu 8/5
• Huainanzi 6, “Lan
<unattributed>故有而若無，實而若虛。
ming 覽冥” (SBCK 7/5a)
曾子曰：以能問於不
能，以多問於寡；有 • Wenzi 3, “Jiu shou 九 老子〔文子〕曰：...故有而若無，實而若
若無，實若虛，犯而 守” (Wenzi shuyi 3.167)
虛。
不校。昔者吾友嘗從
事於斯矣。
Lunyu 8/9
• Guodian “Zun de yi
<unattributed>民可使道之，
尊德義” (Guodian Chu mu 而不可使知之。民可道也，而不可強也。
子曰：民可使由之， zhujian p. 174)
不可使知之。
• Mengzi 7A/5 (SBCK
孟子曰：行之而不著焉，習矣而不察焉，
終身由之而不知其道者，眾也。
13/2b)
Lunyu 8/13
• Da Dai Liji 48, “Bao 明堂之位曰：篤仁而好學，多聞而道慎。
fu 保傅” (SBCK 3/4a)
子曰：篤信好學，守 • Xinshu, “Bao fu 保傅”
(SBCK 5/6b)
死善道。
Lunyu 8/20
...孔子曰：才難，不
其然乎？唐、虞之
際，於斯為盛。有婦
人焉，九人而已。三
分天下有其二，以服
事殷。周之德，
〔其〕可謂至德也已
矣。
Lunyu 8/21
• Yi Zhoushu 64, “Dazi 王子[晉]應之曰：...如文王者，其大道
Jin 大子晉” (Yi Zhoushu 仁，其小道惠，三分天下而有其二，敬人
jiaobu zhuyi 64.402)
無方，服事于商，既有其眾而返失其身，
此之謂仁。
• Shiji 4.51, “Xia benji
夏本紀”
子曰：禹，吾無間然
矣。菲飲食而致孝乎
鬼神，惡衣服而致美
乎黻冕，卑宮室而盡
力乎溝洫。禹，吾無
間然矣。
<unattributed>
禹傷先人父鯀功之不成受誅，
乃勞身焦思，居外十三年，
過家門不敢入。薄衣食，
致孝于鬼神。卑宮室，致費於溝淢。
Lunyu book 9
Lunyu 9/4
子絕四：毋意，毋
必，毋固，毋我。
• Guodian “Yu cong san <unattributed>毋意，毋固，
語叢三” (Guodian Chu mu 毋我，毋必。
zhujian p. 212)
366
• Lüshi chunqiu 13/2,
Lunyu 9/9
子曰：鳳鳥不至，河 “Ying tong 應同” (SBCK
不出圖，吾已矣夫！ 13/5a)
• Guanzi 50, “Feng shan
封禪” (SBCK 16/6a)
• Shiji 28.1361, “Feng
shan shu 封禪書”
• Mozi 52, “Bei cheng
men 備城門” (SBCK
14/1a)
Lunyu 9/16
<unattributed>夫覆巢毀卵，則鳳凰不
至；刳獸食胎，則麒麟不來；乾澤涸漁，
則龜龍不往。
[管仲曰；]...今鳳凰麒麟不來，嘉穀不
生，而蓬蒿藜莠茂，鴟梟數至，而欲封
禪，毋乃不可乎？
禽滑釐問於子墨子曰：由聖人之言，鳳鳥
之不出，諸侯畔殷周之國，甲兵方起於天
下，大攻小，強執弱，吾欲守小國，為之
柰何？
• Mengzi 1A/5 (SBCK
孟子對曰：
...入以事其父兄，出以事其長上。...
子曰：出則事公卿， • Liji, “Xiang yin jiu yi <unattributed>民知尊長養老，
入則事父兄，喪事不 鄉飲酒義” (SBCK 20/6a) 而后乃能入孝弟。民入孝弟，
敢不勉，不為酒困，
出尊長養老，而后成教。
何有於我哉？
1/7b)
Lunyu 9/17
• Mengzi 4B/18 (SBCK
8/6a)
子在川上，曰：逝者
如斯夫！不舍晝夜。
Lunyu 9/23
• Da Dai Liji 49,
“Zengzi li shi” (SBCK
子曰：後生可畏，焉 4/4b)
知來者之不如今也？
四十、五十而無聞
焉，斯亦不足畏也
已。
Lunyu 9/25
• Lüshi chunqiu 20/7,
“Jiao zi 驕恣” (SBCK
子曰：主忠信，毋友 20/18a)
不如己者，過則勿憚
改。
Lunyu 9/26
• Liji, “Ru xing 儒行”
(SBCK 19/5b)
Lunyu 9/28
• Xunzi 27, “Da lüe 大
略” (SBCK 27/17a)
子曰：歲寒，然後知 • Huainanzi 2, “Chu
松柏之後彫也。
zhen 俶真” (SBCK 2/3b)
367
徐子曰：仲尼亟稱於水，曰：水哉，水
哉！何取於水也。孟子曰：（源）〔原〕
泉混混，不舍晝夜。
曾子曰：...三十、四十之閒而無蓺，即無
蓺矣；五十而不以善聞矣；七十而無德，
雖有微過，亦可以勉矣。
李悝趨進曰：昔者楚莊王謀事而當，有大
功，退朝而有憂色。左右曰：王有大功，
退朝而有憂色，敢問其說？王曰：仲虺有
言，不穀說之。曰：諸侯之德，能自為取
師者王，能自取友者存，其所擇而莫如己
者亡。
<unattributed>身可危也。
而志不可奪也。
<unattributed>歲不寒無以知松柏，
事不難無以知君子無日不在是。
<unattributed>夫大寒至，霜雪降，
然後知松柏之茂也。
Lunyu book 10
Lunyu 10/1
Xinyu 1, “Dao ji 道基” <unattributed>...君以仁治，臣以義平，
(SBCK 上/5a)
鄉黨以仁恂恂，朝廷以義便便...
孔子於鄉黨，恂恂如 Shiji 103.2764, “Wan <unattributed>子孫勝冠者在側，
也，似不能言者。
Shi liezhuan 萬石列傳” 雖燕居必冠，申申如也。
其在宗廟朝廷，便便
僮僕訢訢如也，唯謹。
言，唯謹爾。
Shiji 109.2878, “Li
太史公曰：...余睹李將軍悛悛如鄙人，
jiangjun liezhuan 李將 口不能道辭。
軍列傳”
Lunyu 10/2
• Hanshi waizhuan 9
子張曰：子亦聞夫子之議論邪？
徐言誾誾，威儀翼翼。
(SBCK 9/15b)
朝，與下大夫言，侃
侃如也；與上大夫
言，誾誾如也。君
在，踧踖如也，與與
如也。
Lunyu 10/3
• Yili, “Pin li 聘禮”
<unattributed>賓出。公再拜送。
賓不顧。
君召使擯，色勃如
• Yili, “Pin li 聘禮”
<unattributed>賓出。大夫送于外門外。
也，足躩如也。揖所 (SBCK 8/24a)
再拜。賓不顧。
與立，左右手，衣前
• Yili, “Gong shi dafu li <unattributed>賓出。公送于大門內。
後，襜如也。趨進，
公食大夫禮” (SBCK
再拜。賓不顧。
翼如也。賓退，必復 9/9b)
命曰：賓不顧矣。
(SBCK 8/17b)
Lunyu 10/4
• Liji, “Qu li shang 曲
禮上” (SBCK 1/6a)
入公門，鞠躬如也，
如不容。立不中門，
行不履閾。...
• Liji, “Yu zao 玉藻”
<unattributed>大夫士出入君門，
由闑右，不踐閾。
...為人子者，居不主奧，坐不中席，行不
中道，立不中門。
<unattributed>賓入不中門，不履閾。
(SBCK 9/11a)
Lunyu 10/5
• Yili, “Pin li 聘禮”
(SBCK 8/36a)
執圭，鞠躬如也，如 • Liji, “Yu zao 玉藻”
不勝。上如揖，下如 (SBCK 9/11a)
授。勃如戰色，足蹜
蹜如有循。...
Lunyu 10/6
<unattributed>執圭，入門，
鞠躬焉，如恐失之。
<unattributed>執龜玉，舉前曳踵，
蹜蹜如也。
• Liji, “Jian zhuan 間傳 <unattributed>素縞麻衣，中月而禫，
” (SBCK 18/10a)
禫而纖，無所不佩。
...去喪，無所不佩。...
368
Lunyu 10/8
• Hanshi waizhuan 9
...席不正，不坐。...
孟子少時誦，其母方織...其母自悔而言
曰：吾懷妊是子，席不止，不坐；割不
正，不食；胎教之也。...
• Gu lienü zhuan, “Zhou <unattributed>君子謂太任為能胎教。古
shi san mu 周室三母” 者婦人妊子，寢不側，坐不邊，立不蹕，
(SBCK 2/9a)
不食邪味，割不正不食，席不正不坐，
目不視於邪色，耳不聽於淫聲。
• Lunheng 6, “Ming yi <unattributed>性命在本，
命義” (SBCK 2/7b)
故禮有胎教之法：子在身時，
席不正不坐，割不正不食，
非正色目不視，非正聲耳不聽。
Lunyu 10/18
• Liji, “Yu zao 玉藻”
(SBCK 9/1a)
...割不正，不食。...
Lunyu 10/12
<unattributed>若賜之食，而君客之，則
命之祭，然後祭。先飯，
辯嘗羞，飲而俟。
• Yili, “Shi xiang jian li <unattributed>若君賜之食，
士相見禮” (SBCK 3/6a) 則君祭先飯，遍嘗膳，飲而俟。
君命之食，然後食。
(SBCK 13/3a)
君賜食，必正席，先
嘗之。君賜腥，必熟
而薦之。君賜生，必
畜之。侍食於君，君
祭，先飯。
Lunyu 10/20
• Mengzi 2B/2 (SBCK
4/3b)
君命召，不俟駕行
矣。
• Xunzi 27, “Da lüe 大
略” (SBCK 19/1b)
• Liji, “Yu zao 玉藻”
(SBCK 13/8a)
Lunyu 10/25
• Liji, “Yu zao 玉藻”
(SBCK 13/2b)
...迅雷風烈，必變。
Lunyu 10/26
• Liji, “Qu li shang 曲
禮上” (SBCK 1/18a)
升車，必正立，執
綏。車中，不內顧，
不疾言，不親指。
景子曰：否；非此之謂也。禮曰：父召，
無諾；君命召，不俟駕。
<unattributed>諸侯召其臣，臣不俟駕，
顛倒衣裳而走，禮也。
<unattributed>凡君召以三節，
二節以走，一節以趨。
在官不俟屨，在外不俟車。
<unattributed>君子之居恆當戶，
寢恆東首。若有疾風，
迅雷，甚雨，則必變。
<unattributed>國君不乘奇車，
車上不廣欬，不妄指，立視五巂，
式視馬尾，顧不過轂。
Lunyu book 11
[公孫丑]曰：...宰我、子貢善為說辭；冉
牛、閔子、顏淵善言德行。
德行：顏淵、閔子
• Xinxu 3, “Za shi 雜事 孔子曰：言語宰我、子貢。
騫、冉伯牛、仲弓。 ” (SBCK 3/3b)
言語：宰我、子貢。
政事：冉有、季路。
文學：子游、子夏。
Lunyu 11/3
• Mengzi 2A/1 (SBCK
3/8b)
369
Lunyu 11/7
• Lunyu 6/3
哀公問：弟子孰為好學？
孔子對曰：有顏回者好學，
不遷怒，不貳過。不幸短命死矣，
今也則亡，未聞好學者也。
季康子問弟子孰為好
學。孔子對曰：有顏
回者好學，不幸短命
死矣，今也則亡。
Lunyu 11/11
• Shiji 122.3144, “Kuli <unattributed>湯死。
liezhuan 酷吏列傳”
...昆弟諸子欲厚葬湯。
顏淵死，門人欲厚葬
之。
Lunyu 11/16
...子曰：過猶不及。
• Xunzi, “Wang ba 王霸 <unattributed>是過者也，猶不及也。
” (SBCK 11/20b)
• Guanzi 16, “Fa fa 法 <unattributed>過與不及也，皆非正也。
法” (SBCK 6/5b)
Lunyu 11/24
• Guliang zhuan (SBCK
蘧伯玉曰：不以道事其君者其出乎。
9/11a)
...子曰：吾以子為異
之問，曾由與求之
問。所謂大臣者，以
道事君，不可則止。
今由與求也，可謂具
臣矣。
Lunyu book 12
Lunyu 12/1
• Zuozhuan Zhao 12
(SBCK 22/15b)
...子曰：克己復禮為 • Shanghai Museum
仁。...顏淵曰：回雖 “Zhonggong 仲弓” ms.
不敏，請事斯語矣。 strip #9 (Shanghai
仲尼曰：古也有志：克己復禮，仁也。信
善哉！楚靈王若能如是，豈其辱於乾谿。
仲弓曰：雍也不敏。
bowuguan cang Zhanguo
Chu zhushu p. 270)
• Hanshi waizhuan 7
子貢曰：賜雖不敏，請事斯語。
(SBCK 7/15a)
370
Lunyu 12/2
• Zuozhuan Xi 33 (SBCK 臣[冀缺]聞之：出門如賓，承事如祭，仁
7/15a)
之則也。
仲弓問仁。子曰：出 • Mengzi 4A/9 (SBCK
孟子曰：...所欲與之聚之，所惡勿施，爾
門如見大賓，使民如 7/8a)
也。
承大祭。己所不欲，
• Lüshi chunqiu 4/4
<unattributed>所加於人，必可行於己。
勿施於人。在邦無
(SBCK 4/8b)
怨，在家無怨。仲弓
子貢應之曰：君子尊賢而容眾，嘉善而矜
曰：雍雖不敏，請事 • Hanshi waizhuan 9
(SBCK 9/5a)
不能，親內及外，己所不欲，勿施於人。
斯語矣。
• Huainanzi 9, “Zhu shu <unattributed>心之所欲，其不加諸人。
主術” (SBCK 9/21b)
• Shuiyuan 10, “Jing
<unattributed>能行此五者，可以全身，
shen 敬慎” (SBCK
己所不欲，勿施於人，是謂要道也。
10/19b)
• Guanzi 51, “Xiao wen
小問” (SBCK 16/7b)
• Guanzi 66, “Ban fa 版
法” (SBCK 21/4b)
• Shiji 67.2190,
“Zhongni dizi liezhuan
仲尼弟子列傳”
Lunyu 12/4
管子對曰：...非其所欲，勿施於人，仁
也。
<unattributed>己之所不安，勿施於人。
仲弓問政，孔子曰：出門如見大賓，使民
如承大祭。在邦無怨，在家無怨。
• Lüshi chunqiu 14/6,
子路與子貢入。子貢曰：如此者可謂窮
“Shen ren 慎人” (SBCK 矣。孔子曰：...故內省而不疚於道，臨難
14/17b)
司馬牛問君子。子
而不失其德。
•
Zhuangzi
28,
“Rang
曰：君子不憂不懼。
曰：不憂不懼，斯謂 wang 讓王” (SBCK 9/28a)
之君子已乎？
• Liji, “Zhong yong 中 <unattributed>故君子內省不疚，
子曰：內省不疚，夫 庸” (SBCK 16/13a)
無惡於志。君子所不可及者，
何憂何懼？
其唯人之所不見乎。
371
Lunyu 12/5
司馬牛憂曰：人皆有
兄弟，我獨亡。子夏
曰：商聞之矣：死生
有命，富貴在天。君
子敬而無失，與人恭
而有禮。四海之內，
皆兄弟也。君子何患
乎無兄弟也？
• Da Dai Liji 54,
“Zengzi zhi yan 曾子制
言” (SBCK 5/3a)
• Shuiyuan 17, “Za yan
雜言” (SBCK 17/12a)
• Kongzi jiayu 15, “Liu
ben 六本” (SBCK 4/10b)
• Lunheng 3, “Ming lu
命祿” (SBCK 1/10b)
• Lunheng 28, “Wen
Kong 問孔” (SBCK 9/9a)
• Lunheng 72, “Bian
chong 辨崇” (SBCK
曾子曰：...且夫君子執仁立志，
先行後言，千里之外，皆為兄弟，
孔子曰：...賢不肖者才也，為不為者人
也，遇不遇者時也，死生者命也。
孔子曰：...千里之外，親如兄弟。
孔子曰：死生有命，富貴在天。
孔子稱曰：死生有命，富貴在天。
孔子曰：死生有命，富貴在天。
24/12b)
Lunyu 12/6
• Yi Zhoushu, “Shi fa 諡 <unattributed>譖訴不行曰明。
法” (Yi Zhoushu jiaobu
子張問明。子曰：浸 zhuyi 54.289)
潤之譖，膚受之愬，
不行焉，可謂明也已
矣。...
Lunyu 12/11
齊景公問政於孔子。
孔子對曰：君君，臣
臣，父父，子子。公
曰：善哉！信如君不
君，臣不臣，父不
父，子不子，雖有
粟，吾得而食諸？
• Guodian, “Liu de 六 <unattributed>故夫夫，婦婦，
德” ms. (Guodian Chu mu 父父，子子，君君，臣臣，
zhujian pp. 187–88)
此六者各行其職而訕誇蔑由作也。
• Zhouyi #37, “Jiaren 家 <unattributed>父父，子子，
人” (SBCK 4/7b)
兄兄，弟弟，夫夫，婦婦而家道正，
正家而天下定矣。
• Guanzi 2, “Xing shi
君不君，則臣不臣。父不父，則子不子。
形勢” (SBCK 1/5a)
• Guanzi 20, “Xiao
桓公召管仲而謀，管仲對曰：為君不君，
kuang 小匡” (SBCK
為臣不臣，亂之本也。
8/12a)
• Guoyu, “Qi yu 齊語”
(SBCK 6/13a)
• Shuiyuan 6, “Fu en 復 子夏曰：春秋者，記君不君，臣不臣，父
恩” (SBCK 6/20b)
不父，子不子者也
Lunyu 12/14
• Xunzi 32, “Yao wen
堯問” (SBCK 20/24b)
372
堯問於舜曰：我欲致天下，為之奈何？
對曰：執一無失，行微無怠，忠信無倦，
而天下自來。
Lunyu 12/16
• Guliang Yin 1 (SBCK
<unattributed>春秋成人之美。
1/1a)
子曰：
君子成人之美，
不成人之惡。
小人反是。
• Liji, “Ji tong 祭統”
<unattributed>為先祖者，莫不有美焉，
莫不有惡焉，銘之義。稱美而不稱惡，
此孝子孝孫之心也。唯賢者能之。
• Guanzi, “Ban fa 版法” 管子對曰：...故君子惡稱人之惡。
(SBCK 14/22a)
(SBCK 21/7a)
• Da Dai Liji, “Zengzi li 曾子曰：...君子不先人以惡，
shi 曾子立事” (SBCK
不疑人以不信；不說人之過，成人之美。
4/3a)
• Shuiyuan 1, “Jun dao
君道” (SBCK 1/3b)
Lunyu 12/17
哀公曰：善哉！吾聞君子成人之美，不成
人之惡。微孔子，吾焉聞斯言也哉？
• Zuozhuan Huan 2
師服曰：異哉君子之名子也。夫名以制
義，義以出禮，禮以體政，政以正民，是
季康子問政於孔子。
以政成而民聽。
孔子對曰：政者、正 • Guanzi 16, “Fa fa 法 <unattributed>政者，正也；正也者，
也。子帥以正，孰敢 法” (SBCK 6/5b)
所以正定萬物之命也。
不正？
• Da Dai Liji 41, “Ai
[哀]公曰：敢問：何謂為政？孔子對曰：
gong wen yu Kongzi 哀 政者，正也。君為正，則百姓從政矣。...
公問於孔子” (SBCK
(SBCK 2/3b)
1/8a)
• Liji, “Ai gong wen 哀
公問” (SBCK 15/4a)
• Shizi, “Shen ming 神 <unattributed>政也者，正人者也。
明” (SBCK Qunshu zhiyao 身不正則人不從。
36/25a)
373
Lunyu 12/19
• Shangshu, “Jun Chen 王若曰...爾惟風，下民惟草。
君陳” (SBCK /a)
季康子問政於孔子
• Mengzi 3A/2 (SBCK
孟子曰：...君子之德，風也；小人之德，
曰：如殺無道，以就 5/3b)
草也。草上之風，必偃。
有道，何如？孔子對
• Huainanzi 20, “Tai zu 其於化民也，若風之搖草木，無之而不
曰：子為政，焉用
xun 泰族訓” (SBCK
靡。
殺？子欲善而民善
20/11a)
矣。君子之德風，小
• Huainanzi 10, “Miu <unattributed>人能尊道行義，
人之德草。草上之
cheng 繆稱” (SBCK 10b– 喜怒取予，欲如草之從風。
風，必偃。
11a)
• Wenzi 2, “Jing cheng
精誠” (Wenzi shuyi 2.105)
• Xinyu 4, “Wu wei 無
為” (SBCK 上/11a)
• Guoyu, “Jin yu 晉語”
文子曰：...大人以善示人，不變其故，
不易其常，天下聽令，如草從風。
故上之化下，猶風之靡草也。
公子縶曰：...殺無道而立有道，仁也。
(SBCK 9/7a)
• Shuiyuan 1, “Jun dao
君道” (SBCK 1/2a)
• Shuiyuan 5, “Gui de
貴德” (SBCK 5/15a–b)
Lunyu 12/22
陳靈公行僻而言失。泄冶曰：...夫上之化
下，猶風靡草，東風則草靡而西，西風則
草靡而東，在風所由而草為之靡。
<unattributed>上之變下，猶風之靡草
也。
• Xunzi 29, “Zi dao 子
道” (SBCK 20/13a)
子貢入，子曰：賜！知者若何？仁者若
何？子貢對曰：知者知人，仁者愛人。
樊遲問仁。子曰：愛
子曰：可謂士君子矣。
人。問知。子曰：知 • Huainanzi 20, “Tai zu <unattributed>故仁知，人材之美者也。
人。樊遲未達。
xun 泰族訓” (SBCK
所謂仁者，愛人也；所謂知者，知人也。
子曰：舉直錯諸枉， 20/18b)
能使枉者直。...
• Wenzi 4, “Fu yan 符言 文子曰：舉枉與直，如何不得；舉直與
” (Wenzi yishu 4.213)
枉，勿與遂往；所謂同污而異泥者。
Lunyu book 13
374
Lunyu 13/3
...子曰：必也正名
乎！子路曰：有是
哉，子之迂也！奚其
正？子曰：野哉由
也！君子於其所不
知，蓋闕如也。名不
正，則言不順；言不
順，則事不成；事不
成，則禮樂不興；禮
樂不興，則刑罰不
中；刑罰不中，則民
無所錯手足。故君子
名之必可言也，言之
必可行也。君子於其
言，無所苟而已矣。
• Lüshi chunqiu 17/1,
<unattributed>故至治之務，在於正名。
“Shen fen 審分” (SBCK 名正則人主不憂勞矣，
17/3a)
不憂勞則不傷其耳目之主。
Lunyu 13/7
• Shiji 65.2165, “Sunzi 魯人或惡吳起曰：...且魯衛兄弟之國也。
Wu Qi liezhuan 孫子吳
子曰：魯、衛之政， 起列傳”
兄弟也。
Lunyu 13/10
• Han Feizi (SBCK /a)
子曰：苟有用我者，
期月而已可也，三年 • Han Feizi (SBCK /a)
有成。
• Han Feizi (SBCK /a)
<unattributed>宋人有為其君以象為楮葉
者，三年而成。
<unattributed>墨子為木鳶，三年而成...
<unattributed>客有為周君畫莢者，
三年而成...
<unattributed>齊景公使人為弓，
三年乃成...
• Hanshi waizhuan 8
(SBCK 8/15b)
Lunyu 13/15
• Mao Shi #236 (SBCK
<unattributed>明明在下，赫赫在上。
天難忱斯，不易維王。
...孔子對曰：言不可 • Shuiyuan 11, “Shan
公乘不仁曰：...為人臣者不易，
以若是其幾也。人之 shui 善說” (SBCK 11/10b) 為君亦不易。
言曰：為君難，為臣
不易。
/a)
Lunyu 13/16
...子曰：近者說，遠
者來。
• Xunzi 27, “Da lüe 大
略” (SBCK 19/18a)
曾子曰：孝子言為可聞，行為可見。言為
可聞，所以說遠也；行為可見，所以說近
也；近者說則親，遠者悅則附；親近而附
遠，孝子之道也。
• Guanzi 64, “Xing shi <unattributed>明主之使遠者來
形勢” (SBCK 20/5b)
而近者親也，為之在心
• Wenzi 7, “Wei ming 老子〔文子〕曰：古者親近不以言，來遠
微明” (Wenzi yishu 7.336) 不以言，使近者悅，遠者來。
375
Lunyu 13/18
葉公語孔子曰：吾黨
有直躬者，其父攘
羊，而子證之。孔子
曰：吾黨之直者異於
是：父為子隱，子為
父隱。直在其中矣。
Lunyu 13/20
• Lüshi chunqiu 11/4,
楚有直躬者，其父竊羊而謁之上，上執而
“Dang wu 當務” (SBCK 將誅之。直躬者請代之。將誅矣，告吏
11/8a–b)
曰：父竊羊而謁之，不亦信乎？父誅而代
之，不亦孝乎？信且孝而誅之，國將有不
誅者乎？荊王聞之，乃不誅也。孔子聞之
曰：異哉直躬之為信也，一父而載取名
焉。
• Mozi 16, “Jian ai xia
兼愛下” (SBCK 4/11a &
當使若二士者，言必信，行必果，使言行
之合猶合符節也，無言而不行也。
...曰：言必信，行必 4/12b)
果，硜硜然小人哉！ • Mengzi 4B/11 (SBCK
8/4b)
抑亦可以為次矣。
孟子曰：大人者，言不必信，行不必果，
惟義所在。
• Shiji 124.3181,
今游俠，其行雖不軌於正義，
“Youxia liezhuan 游俠 然其言必信，其行必果。
列傳”
Lunyu 13/25
子曰：君子易事而難
說也。說之不以道，
不說也；及其使人
也，器之。小人難事
而易說也。說之雖不
以道，說也；及其使
人也，求備焉。
• Shanghai Museum
(甲17)〔君子先〕人則啟道之，後人則奉
“Cong zheng 從政” ms. 相之，是以曰君子難得而易使也。其使
人，器之。小人先人則弁敔之。〔後
人〕(甲18)則暴毀之，是以曰小人易得而
難使也。其使人必，求備焉。
• Xunzi 27, “Da lüe 大 <unattributed>故曰：君子難說，說之不
略” (SBCK 19/24a)
以道，不說也。
Lunyu 13/28
• Mao Shi, commentary
to #164, “Chang di 常
子曰：切切偲偲，怡 棣”
怡如也，可謂士矣。 • Da Dai Liji 49,
朋友切切偲偲，兄弟 “Zengzi lishi 曾子立事
怡怡。
” (SBCK 4/7a)
<unattributed>兄弟尚恩怡怡然，
朋友以義切切然。
Lunyu 13/29
子墨子言曰：...古者吳闔閭教七年，奉甲
執兵，
• Mozi 18, “Fei gong
zhong 非攻中” (SBCK
5/5b–6a)
子曰：善人教民七
年，亦可以即戎矣。 • Mawangdui “Jun
zheng 君正” ms.
宮中雍雍，外焉肅肅，兄弟僖僖，朋友切
切，遠者以貌，近者以情。
<unattributed>一年從其俗，
二年用其德，三年而民有得，
四年而發號令，五年而以刑正，
六年而民畏敬，七年而可以正（征）。
376
Lunyu 13/30
• Mengzi 6B/8 (SBCK
12/11a)
子曰：以不教民戰， • Guliang Xi 23 (SBCK
是謂棄之。
5/19a)
魯欲使慎子為將軍。孟子曰：不教民而用
之謂之殃民。...
<unattributed>以其不教民戰，則是棄其
師也。為人君而棄其師，
其民孰以為君哉。
Lunyu book 14
Lunyu 14/1
憲問恥。子曰：邦有
道、穀，邦無道、
穀，恥也。克、伐、
怨、欲不行焉，可以
為仁矣？子曰：可以
為難矣，仁則吾不知
也。
Lunyu 14/10
• Zuozhuan Zhao 1
(SBCK 20/7b)
• Guoyu (SBCK 14/12a)
• Yanzi chunqiu (SBCK
6/16a)
Lunyu 14/16
子路曰：桓公殺公子
糾，召忽死之，管仲
不死。曰：未仁乎？
子曰：桓公九合諸
侯，不以兵車，管仲
之力也。如其仁，如
其仁。
鍼[=秦后子]聞之：國無道而年穀和熟，
天贊之也。鮮不五稔。
...晏子辭不受，曰：富而不驕者，未嘗聞
之。貧而不恨者，嬰是也。所以貧而不恨
者，以善為師也。
• Lüshi chunqiu 17/4,
<unattributed>[齊桓公]十年，九合諸
“Wu gong 勿躬” (SBCK 侯，一匡天下，皆夷吾與五子之能也。
17/11b)
管子，人臣也...
• Xunzi 11, “Wang ba <unattributed>齊桓公...九合諸侯，
王霸” (SBCK 7/19b)
一匡天下，為五伯長...知一政於管仲也，
是君人者之要守也。
• Hanshi waizhuan 6
<unattributed>桓公之所以九合諸侯，
(SBCK 6/1b)
一匡天下者、此也。
• Han Feizi 10, “Shi
<unattributed>昔者齊桓公九合諸侯，
guo 十過” (SBCK 3/4b) 一匡天下，為五伯長，管仲佐之。
• Han Feizi 35, “Wai
[蘇代]曰：昔者齊桓公愛管仲，置以為仲
chushuo you 外儲説右 父，內事理焉，外事斷焉，舉國而歸之，
” (SBCK 14/5a)
故一匡天下，九合諸侯。
• Han Feizi 37, “Nan
晉平公問叔向曰：昔者齊桓公九合諸侯，
san 難三” (SBCK 15/8a) 一匡天下，不識臣之力也？君之力也
• Zhanguo ce (SBCK
蔡澤曰：昔者，齊桓公九合諸侯，
8/36b)
一匡天下
• Huainanzi 13, “Fan
<unattributed>然而管仲免於累紲之中，
lun 氾論” (SBCK 13/14a) 立齊國之政，九合諸侯，一匡天下。
• Shiji 62.2131, “Guan 齊桓公以霸，九合諸侯，
Yan liezhuan 管晏列傳 一匡天下，管仲之謀也。
”
377
Lunyu 14/24
• Xunzi 1,
<unattributed>古之學者為己，
“Quan xue 勸學” (SBCK 今之學者為人。
1/6b)
子曰：古之學者為
己，今之學者為人。 • Xinxu (Quan Han wen
39/1a; p. 340)
Lunyu 14/26
齊王問墨子曰：古之學者為己，
今之學者為人，何如？
• Zhouyi #52, “Liang 良 象曰：兼山，艮。君子以思不出其位。
” (SBCK 5/13a)
子曰：不在其位，不
謀其政。曾子曰：君
子思不出其位。
Lunyu 14/27
• Liji, “Za ji 雜記”
<unattributed>有其言，無其行，
君子恥之。
<unattributed>君子有五恥：
...有其言，無其行，君子恥之。
(SBCK 12/16b)
子曰：君子恥其言而 • Shuiyuan 16, “Tan
過其行。
cong 談叢” (SBCK
16/15a)
Lunyu 14/35
• Laozi 70 (SBCK 下/16b) <unattributed>吾言甚易知，甚易行。
天下莫能知，莫能行。言有宗，事有君。
子曰：莫我知也夫！
夫唯無知，是以不我知。知我者希，
子貢曰：何為其莫知
則我者貴，是以聖人被褐懷玉。
子也？
• Chuci, “Li sao 離騷” [屈原]亂曰：已矣哉，國無人莫我知兮。
子曰：不怨天，不尤 (SBCK 1/49a)
人，下學而上達。知 • Mengzi 2B/13 (SBCK 孟子去齊，充虞路問曰：夫子若〔有〕不
我者其天乎！
4/15b)
豫色然。前日虞聞諸夫子曰：君子不怨
• Xunzi 4, “Rong ru 榮
辱” (SBCK 2/14a)
• Huainanzi 10, “Miu
cheng 繆稱” (SBCK
天，不尤人。
<unattributed>自知者不怨人，
知命者不怨天。
<unattributed>是故知己者不怨人，
知命者不怨天。
10/13a)
• Shiji 84.2494, “Qu
[賈誼]訊曰：已矣，國其莫我知。
Yuan Jia Sheng
liezhuan 屈原賈生列傳
”
• Liji, “Zhong yong 中 <unattributed>君子無入而不自得焉。
庸” (SBCK 16/4a) (SBCK 在上位不陵下，在下位不援上，
/a)
正己而不求於人，則無怨。上不怨天，
下不尤人，故君子居易以俟命。
Lunyu 14/38
• Gongyang Xuan 8
<unattributed>存其心焉爾者何？
知其不可而為之也。
(SBCK 7/7b)
曰：是知其不可而為
之者與？
378
Lunyu 14/40
• Shangshu, “Yi xun 伊 <unattributed>百官總己以聽冢宰。
訓” (SBCK 4/6a)
...子曰：何必高宗，
古之人皆然。君薨，
百官總己以聽於冢宰
三年。
Lunyu 14/42
• Hanshi waizhuan 7
狐丘丈人曰：善哉！言乎！堯舜其猶病
諸！
魏文侯...曰：善哉言乎！堯舜其猶病諸！
(SBCK 7/9a)
• Hanshi waizhuan 8
(SBCK 8/19a)
Lunyu book 15
Lunyu 15/1
• Zuozhuan Ai 11 (SBCK 孔文子之將攻大叔也。訪於仲尼。
29/21a)
仲尼曰：胡簋之事，則嘗學之矣。
衛靈公問陳於孔子。
甲兵之事，未之聞也。
孔子對曰：俎豆之
事，則嘗聞之矣；軍
旅之事，未之學也。
明日遂行。
Lunyu 15/18
子曰：君子義以為
質，禮以行之，孫以
出之，信以成之。君
子哉！
Lunyu 15/21
• Xinyu 6, “Shen wei 慎 <unattributed>顏回一簞食，一瓢飲，
微” (SBCK 上/15a)
在陋巷之中，人不堪其憂，
回也不改其樂。禮以行之，遜以出之。
• Guanzi 31, “Jun chen <unattributed>所求於己者多，
xia 君臣下” (SBCK
故德行立。所求於人者少，故民輕給之。
子曰：君子求諸己， 11/4b)
小人求諸人。
• Huainanzi 10, “Miu
cheng xun 繆稱訓”
<unattributed>故怨人不如自怨，
求諸人不如求諸己得也。
(SBCK 10/5b)
• Wenzi 6, “Shang de 上 老子〔文子〕曰：...故怨人不如怨己，免
德 (Wenzi shuyi 6.300)
求諸人不如求諸己。
Lunyu 15/24
<see 12/2>
379
Lunyu 15/27
• Guodian “Yu cong er <unattributed>小不忍，敗大勢。
語叢二” (Guodian Chu mu
子曰：...小不忍，則
亂大謀。
zhujian p. 205)
• Preface to Mao Shi
<unattributed>小不忍以致大亂焉。
#76, “Qiang Zhongzi 將
仲子” (SBCK 4/8b)
• Shiji 58.2091–92,
<unattributed>小不忍害大義狀報太后。
“Liang Xiaowang
liezhuan 梁孝王列傳”
380
Lunyu 15/30
• Zhouyi #42, “Yi 益”
(SBCK 4/14a)
子曰：過而不改，
是謂過矣。
• Shanghai Museum
“You huang jiang qi 有
皇將起” (Shanghai
<unattributed>象曰：風雷，益。
君子以見善則遷，有過則改。
<unattributed>有過而能改。
bowuguan cang Zhanguo
Chu zhushu vol. 8, p. 275)
• Zuozhuan Xuan 2
(SBCK 10/3b)
• Chuci, “Tian wen 天
文” (SBCK 3/34a)
• Mengzi 2B/9 (SBCK
4/11b)
• Mengzi 6B/19 (SBCK
[晉靈公]曰：吾知所過矣，將改之。[士
季]稽首而對曰：人誰無過，過而能改，
善莫大焉。詩曰：靡不有初，鮮克有終。
夫如是，則能補過者鮮矣。君能有終，則
社稷之固也。豈惟群臣賴之？又曰：袞職
有闕，惟仲山甫補之，能補過也。君能補
過，袞不廢矣。
<unattributed>悟過改更，我又何言？
孟子曰：...且古之君子，過則改之；
今之君子，過則順之。
古之君子，其過也，如日月之食，
民皆見之；及其更也，民皆仰之。
今之君子，豈徒順之，又從為之辭。
孟子曰：...人恆過，然後能改
12/17a)
• Guoyu, “Lu yu shang 文子聞之，曰：過而能改者，民之上也。
魯語上” (SBCK 4/16a)
• Guanzi 31, “Jun chen <unattributed>君有過而不改，謂之倒。
xia 君陳下” (SBCK
11/2b)
• Guliang Xi 22 (SBCK
5/18a)
• Da Dai Liji, “Zengzi
lishi 曾子立事” (SBCK
<unattributed>過而不改，又之是謂之
過。
曾子曰 ：...過而不能改，倦也。
行而不能遂，恥也。...
4/4b–5a)
• Da Dai Liji 66,
<unattributed>過，失也。
“Sheng de 盛德” (SBCK 人情莫不有過，過而改之，是不過也。
8/9b)
• Shuiyuan 1, “Jun dao
君道” (SBCK 1/17a)
381
君子曰：夫過而改之，是猶不過。
Lunyu 15/40
• Xunzi 27, “Da lüe 大
略” (SBCK 19/22b–23a)
<unattributed>君人者不可以不慎取臣，
匹夫不可不慎取友。友者、所以相有也。
子曰：道不同，不相
道不同，何以相有也？
為謀。
• Shiji 63.2143, “Laozi <unattributed>世之學老子者則絀儒學，
Han Fei liezhuan 老子 儒學亦絀老子。道不同不相為謀，豈謂是
韓非列傳”
邪？
Lunyu 15/41
• Yili, “Pin li 聘禮”
<unattributed>辭苟足以達。義之至也。
(SBCK 8/34b)
子曰：辭達而已矣。
Lunyu book 16
Lunyu 16/2
孔子曰：天下有道，
則禮樂征伐自天子
出；天下無道，則禮
樂征伐自諸侯出。自
諸侯出，蓋十世希不
失矣；自大夫出，五
世希不失矣；陪臣執
國命，三世希不失
矣。天下有道，則政
不在大夫。天下有
道，則庶人不議。
Lunyu 16/6
• Kong congzi 18, “Wen 陳王問大（太）師曰：
jun li 問軍禮” (SBCK
行軍之禮、可得備聞乎？答曰：
6/12a)
天下有道，禮樂、征伐自天子出。
• Shuiyuan 19, “Xiu
<unattributed>天下有道，
wen 修文” (SBCK 19/1a) 則禮樂征伐自天子出。
• Mao commentary to <unattributed>天下有道，
Mao Shi #133, “Wu yi 則禮樂征伐自天子出。
無衣” (SBCK 6/15a)
• Hanshi waizhuan 4
<unattributed>故未可與言而言，
謂之瞽，可與言而不與言，謂之隱，
君子不瞽，言謹其序。
• Xunzi 1, “Quan xue 勸 <unattributed>故未可與言而言，
學” (SBCK 1/8b)
謂之傲；可與言而不言，謂之隱；
不觀氣色而言，謂瞽。
故君子不傲、不隱、不瞽，謹順其身。
(SBCK 4/9a)
孔子曰：侍於君子有
三愆：言未及之而言
謂之躁，言及之而不
言謂之隱，未見顏色
而言謂之瞽。
Lunyu 16/11
• Huainanzi 10, “Miu
cheng 繆稱” (SBCK
孔子曰：見善如不
及，見不善如探
湯。...
10/5a–b)
Lunyu 16/14
• Zuozhuan Zhao 7
<unattributed>文王聞善如不及，
宿不善如不祥。
• Wenzi 6, “Shang de 上 文子曰：...故見善如不及，
德” (Wenzi shuyi 6.300)
宿不善如不祥。
九月，公至自楚。孟僖子病不能相禮，乃
講學之。苟能禮者從之，及其將死也。召
其大夫曰：禮，人之幹也。無禮無以立。
吾聞將有達者，曰孔丘，聖人之後也。
(SBCK 21/20a)
[子曰：]不學禮，無
以立。
Lunyu book 17
382
Lunyu 17/1
• Hanshi waizhuan 1
(SBCK 1/1a)
陽貨欲見孔子...曰： • Chuci, “Li sao 離騷”
「懷其寶而迷其邦， (SBCK 1/6b)
可謂仁乎？
Lunyu 17/6
• Mengzi 4A/16 (SBCK
<unattributed>曾子重其身而輕其祿。
懷其寶而迷其國者，不可與語仁。
[屈原] 汩余若將不及兮，
恐年歲之不吾與。
孟子曰：恭者不侮人...
7/12a)
孔子曰：...恭、寬、
信、敏、惠。恭則不
侮...
Lunyu 17/10
• Shangshu Kong An- <unattributed>人而不學，其猶正牆面而
guo commentary (SBCK 立，臨政事必煩。
子謂伯魚曰：女為周 1/3a)
南、召南矣乎？人而
不為周南、召南，其
猶正牆面而立也與？
Lunyu 17/11
• Xunzi 27, “Da lüe 大
略” (SBCK 19/3a)
子曰：禮云禮云，玉 • Shiji 24.1196, “Yue
帛云乎哉？樂云樂
shu 樂書” (SBCK /a)
云，鍾鼓云乎哉？
聘禮志曰：幣厚則傷德，財侈則殄禮。
禮云禮云，玉帛云乎哉！
故聖人曰：禮云樂云。
Lunyu 17/17
<see 1/3>
子曰：巧言令色，鮮
矣仁。
Lunyu 17/19
...子曰：天何言哉？
四時行焉，百物生
焉，天何言哉？
Lunyu 17/24
...子曰：有惡：惡稱
人之惡者，惡居下流
而訕上者，惡勇而無
禮者，惡果敢而窒
者。...
• Xunzi 3, “Bu gou 不苟 <unattributed>天不言而人推其高焉，
” (SBCK 2/6a)
地不言而人推其厚焉，
四時不言而百姓期焉。
• Mengzi 5A/5 (SBCK
孟子曰：天不言，以行與事示之而已矣。
9/9a)
• Guanzi 66, “Ban fa 版 管子對曰：...故君子惡稱人之惡，
法” (SBCK 21/7a)
惡不忠而怨妒，惡不公議而名當稱，
惡不位下而位上，惡不親外而內放，
Lunyu book 18
383
Lunyu 18/6
• Huainanzi 9, “Zhu shu <unattributed>夫鳥獸之不可同群者，
xun 主術訓” (SBCK 9/8b) 其類異也。
...夫子憮然曰：
鳥獸不可與同群，
吾非斯人之徒與
而誰與？...
Lunyu 18/7
...子路曰：不仕無
義。長幼之節，不可
廢也；君臣之義，如
之何其廢之？欲絜其
身，而亂大倫。君子
之仕也，行其義也。
道之不行，已知之
矣。
Lunyu 18/8
Xinyu 6, “Shen wei 慎
微” (SBCK 上/15b)
道者，人之所行也。夫大道履之而行，
則無不能，故謂之道。
故孔子曰：道之不行也。
Liji, “Zhong yong 中庸 子曰：道之不行也，我知之矣。
” (SBCK 16/1b)
Shiji 130.3297, “Taishi 太史公曰：余聞董生曰：周道衰廢，
gong xu 太史公序”
孔子為魯司寇，諸侯害之，大夫壅之。
孔子知言之不用，道之不行也。
• Hanshi waizhuan 5
<unattributed>言中倫，行中理，
天下順矣。
(SBCK 5/7b)
子曰：不降其志，不
辱其身，伯夷、叔齊
與！
謂柳下惠、少連，降
志辱身矣，言中倫，
行中慮，其斯而已
矣。
Lunyu book 19
Lunyu 19/1
• Shanghai Museum
孔子曰：夫祭，至敬之本也，
“Zhonggong 仲弓” ms., 所以立生也，不可不慎也，
子張曰：士見危致
strips #6 & 23 (Shanghai 夫喪至愛之卒也，所以成死也，
命，見得思義，祭思 bowuguan cang Zhanguo
不可不慎也
Chu
zhushu,
vol.
3,
pp.
267
敬，喪思哀，其可已
& 279–280
矣。
• Zuozhuan Wen 15
齊人送之書曰：...史佚有言曰：兄弟致
(SBCK 9/8a)
美，救乏，賀善，弔災，祭敬，喪哀。情
雖不同，毋絕其愛親之道也。
• Zhuangzi 31, “Yu fu 孔子愀然曰：請問何謂真？客曰：...處喪
漁父” (SBCK 10/9b)
以哀為主。...
• Liji, “Shao yi 少儀” <unattributed>賓客主恭，祭祀主敬，
(SBCK 10/15b)
喪事主哀，會同主詡。
384
Lunyu 19/3
• Hanshi waizhuan 9
(SBCK 9/5a)
...子張曰：異乎吾所
聞：君子尊賢而容
眾，嘉善而矜不能。
我之大賢與，於人何
所不容？我之不賢
與，人將拒我，如之
何其拒人也？
• Liji, “Ru xing 儒行”
(SBCK 19/6a)
• Liji, “Zhong yong 中
庸” (SBCK 16/8a)
...子貢應之曰：君子尊賢而容眾，嘉善而
矜不能。...
<unattributed>儒有博學而不窮，
篤行而不倦，幽居而不淫，上通而不困。
禮之以和為貴，忠信之美，
優游之法，慕賢而容眾。
<unattributed>嘉善而矜不能。
• Kongzi jiayu 17, “Ai 孔子曰...嘉善而矜不能。
gong wen zheng 哀公
問政” (SBCK 4/19a)
• Kongzi jiayu 5, “Ru
孔子侍坐曰：...慕賢而容眾。
xing 儒行” (SBCK 1/17b–
18a)
Lunyu 19/8
• Shiji 47.1916,
“Kongzi shijia 孔子世
子夏曰：小人之過也 家”
必文。
• He Xiu 何休 commentary to Gongyang
Ding 10 (SBCK 11/9a)
Lunyu 19/11
有司進對曰：君子有過則謝以質，小人有
過則謝以文。...
晏子曰：君子謝過以質，小人謝過以文。
• Hanshi waizhuan 2
孔子曰：...且夫齊程本子，天下之賢士
也，吾於是不贈，終身不之見也。
大德不踰閑，小德出入可也。
• Shuiyuan 8, “Zun xian 孔子曰：...今程子天下之賢士也，
尊賢” (SBCK 8/21a)
於是不贈，終身不見。
大德毋踰閑，小德出入可也。
• Yanzi chunqiu (SBCK 晏子曰：...且吾聞之，
5/17a–b)
大者不踰閑，小者出入可也。
晏子出，仲尼送之以賓客之禮。
(SBCK 2/10b)
子夏曰：
大德不踰閑，
小德出入可也。
Lunyu 19/14
• Xiaojing 10, “Ji xiao
xing 紀孝行” (SBCK
子游曰：喪致乎哀而 1/9a)
止。
Lunyu 19/19
子曰：孝子之事親也。...喪則致其哀。
<see also 19/1>
• Shangshu dazhuan
子曰：聽訟者，雖得其情，必哀矜之。
(SBCK 4/24b)
孟氏使陽膚為士師， • Hanshi waizhuan 3
問於曾子。曾子曰： (SBCK 3/12b)
上失其道，民散久
矣。如得其情，則哀
矜而勿喜！
傳曰：魯有父子訟者、康子欲殺。
孔子曰：未可殺也。
夫民父子訟之為不義久矣，
是則上失其道，上有道，是人亡矣。
385
Lunyu 19/21
• Mengzi 2B/9 (SBCK
孟子曰：...且古之君子，過則改之；
今之君子，過則順之。古之君子，
其過也，如日月之食，民皆見之；
及其更也，民皆仰之。
• Huainanzi 20, “Tai zu 故君子之過也，猶日月之蝕，何害於明！
泰族” (SBCK 20/12a)
• Wenzi 11, “Shang yi 老子〔文子〕曰：...夫君子之過，獨日月
上義” (Wenzi shuyi 11.487) 之觸。
• Shuiyuan 16, “Tan
<unattributed>君子之過猶日月之蝕也，
cong 談叢” (SBCK
何害於明？
4/11b)
子貢曰：
君子之過也，
如日月之食焉：
過也，人皆見之；
更也，人皆仰之。
16/16b)
• Kong congzi 12, “Ru
fu 儒服” (SBCK 4/11b)
Lunyu 19/22
衛公孫朝問於子貢
曰：仲尼焉學？子貢
曰：文武之道，未墜
於地，在人。賢者識
其大者，不賢者識其
小者。莫不有文武之
道焉。夫子焉不學？
而亦何常師之有？
子高曰：...夫君子之敗，如日月之蝕。
• Shiji 67.2196,
陳子禽問子貢曰：「仲尼焉學？」子貢
“Zhongni dizi liezhuan 曰：「文武之道未墜於地，在人，賢者識
仲尼弟子列傳”
其大者，不賢者識其小者，莫不有文武之
道。夫子焉不學，而亦何常師之有！」又
問曰：「孔子適是國必聞其政。求之與？
抑與之與？」子貢曰：「夫子溫良恭儉讓
以得之。夫子之求之也，其諸異乎人之求
之也。」
Lunyu book 20
Lunyu 20/2
...子曰：不教而殺謂
之虐，不戒視成謂之
暴，慢令致期謂之
賊，猶之與人也，出
納之吝謂之有司。
Lunyu 20/3
• Shanghai Museum
<unattributed>不修不武，
“Cong zheng 從政”
謂之必成則暴，
ms., strip # 15 (Shanghai 不教而殺則號，令無時。
bowuguan cang Zhanguo
Chu zhushu, vol. 2, p. 228)
• Hanshi waizhuan 3
子貢曰：...賜聞之：託法而治，謂之暴；
不戒致期，謂之虐；不教而誅，謂之賊；
以身勝人，謂之責。
• Shuiyuan 16, “Tan
<unattributed>不教而誅謂之虐，
cong 談叢” (SBCK 16/2a) 不戒責成謂之暴也。
(SBCK 3/15b)
• Zuozhuan Zhao 7
孟僖子病不能相禮，乃講學之。
苟能禮者從之，及其將死也。召其大夫
曰：禮，人之幹也。無禮無以立。
(SBCK 21/20a)
孔子曰：不知命，無
以為君子也；不知
禮，無以立也；不知
言，無以知人也。
386
CHAPTER THREE APPENDICES
• Appendix A: Emperor Wen’s 165 BCE edict (Hanshu 49.2290)
In the fifteenth year [of Emperor Wen’s reign] on the day renzi in the ninth month, the Emperor
said, “Formerly the Great Yu was diligent in his search for men of accomplishment, reaching beyond his realm to everywhere within the four poles. Wherever boats and chariots reached or humans trod there were none who did not hear of his mandate [to find] those who would assist him
with his shortcomings. Those near at hand offered up their discernment and those far away contributed their perspicacity as they came together to assist the Son of Heaven. This is why the
Great Yu was able to avoid losing his virtue and the Xia flourished for so long. Emperor Gaozu
himself rid [the world] of great harm and all traces of disorder, establishing extraordinary heroes
as officials and leaders. They remonstrated with and countered him, assisting the Son of Heaven
with his inadequacies and supporting the Han clan. [With their help he] relied on the numenous
power of heaven, he earned the good fortune of the ancestral temple, he pacified his realm, and
his beneficence spread as far as the barbarians of the four directions.
Now We have won possession of the Son of Heaven’s rule and received the sacrifices of the
ancestral temple. We are both lacking in virtue and in intelligence. We are not even as bright as a
candle, nor is our knowledge up to the task of ruling. This the grandees already know well.
Thus I order the responsible officials, the various feudal kings, the Three Excellencies, the
Nine Ministers as well as those in charge of officers in the commanderies to each strive to the utmost to select men of accomplishment and integrity who understand the great corporate body of
the state, who comprehend the beginnings and endings of human affairs, and who can speak
frankly and remonstrate. Let men be selected according to the population of each province who
will correct Our shortcomings.
We will greatly approve of those grandees whose conduct matches this three-fold Way, and
we will bring those grandees to court to personally instruct us as to Our [proper] aims. Let the
grandees submit [memorials on] the essentials of the three-fold way and also thoroughly contemplate Our lack of virtue, the injustices perpetrated by Our officers, the mistakes of Our government, and disturbances to Our people. Let all [the recommended grandees] display their aims
without keeping anything hidden. Above let them promote the former emperors’ ancestral temples, below let them stimulate the people to virtue and benefit.
Let them compose [their responses] on bamboo strips, which We will personally survey in
order to discern how the grandees might assist Us and whether or not they meet [Our criteria].
Write comprehensively, meticulously, solemnly, and completely. We Ourself promote this. Let
the grandees rectify their discussions without impairing their responsibilities. Be warned! Let the
grandees exert themselves to the fullest without being negligent.
387
惟十有五年九月壬子，皇帝曰：昔者大禹勤求賢士，施及方外，四極之內，舟車所至，人
跡所及，靡不聞命，以輔其不逮；近者獻其明，遠者通厥聰，比善戮力，以翼天子。是以
大禹能亡失德，夏以長楙。高皇帝親除大害，去亂從，並建豪英，以為官師，為諫爭，輔
天子之闕，而翼戴漢宗也。賴天之靈，宗廟之福，方內以安，澤及四夷。
今朕獲執天子之正，以承宗廟之祀，朕既不德，又不敏，明弗能燭，而智不能治，此
大夫之所著聞也。
故詔有司、諸侯王、三公、九卿及主郡吏，各帥其志，以選賢良明於國家之大體，通
於人事之終始，及能直言極諫者，各有人數，將以匡朕之不逮。二三大夫之行當此三道，
朕甚嘉之，故登大夫于朝，親諭朕志。大夫其上三道之要，及永惟朕之不德，吏之不平，
政之不宣，民之不寧，四者之闕，悉陳其志，毋有所隱。上以薦先帝之宗廟，下以興愚民
之休利，著之于篇，朕親覽焉，觀大夫所以佐朕，至與不至。書之，周之密之，重之閉
之。興自朕躬，大夫其正論，毋枉執事。烏虖，戒之！二三大夫其帥志毋怠！
388
• Appendix B: Lunyu 5/26 & 11/26
Lunyu 5/26:
When Yan Yuan and Zilu were attending to him the Master said, “Why don’t you each speak of
your aims?”
Zilu said, “I wish to share my carriage, horse, clothing, and furs with my friends, and I will
not regret it should they become worn out.”
Yan Yuan said, “I wish never to boast of my virtues nor to impose burdens on others.”
Zilu said, “I wish to hear your aims, Master.” The Master said, “To comfort the elderly, to
trust in my friends, and to cherish the young.”
顏淵季路侍。子曰：盍各言爾志？子路曰：願車馬衣（輕）裘與朋友共，敝之而無憾。顏
淵曰：願無伐善，無施勞。子路曰：願聞子之志。子曰：老者安之，朋友信之，少者懷
之。
Lunyu 11/26
When Zilu, Zengxi, Ran You, and Gongxi Hua were sitting in attendance the Master said, “Do
not mind me just because I am somewhat older than you. You are all in the habit of saying, “I am
not recognized.” If someone was to recognize [your abilities], what would you do?”
Zilu was the first to answer: “If I were to govern a state of a thousand chariots stuck between
two great states that suffered armed invasions and repeated famines, I could, within three years,
give its people courage and a sense of direction.”
The Master smiled at him. “Qiu, how about you?”
Ran Qiu answered, “If I were to govern a state of sixty or seventy square li, or even fifty or
sixty square li, I could, within three years, bring the size of the population up to an adequate level. As for ritual and music, I would leave that to a noble man.”
“Chi, how about you?”
Gongxi Hua said, “I am not saying that I could do this, just that I wish to learn. In service of
the ancestral temple or at diplomatic gatherings I should like to wear a ceremonial cap and gown
and assist as a minor official.”
“Dian, what about you?”
He stopped playing his qin and set it down with a clatter before rising to answer, “My answer
differs from all the others.”
The Master said, “What harm is there in that? After all, you are each speaking of your aims.”
389
Zeng Xi said, “In late spring when the spring clothes are newly made, with five or six adults
and six or seven boys I should like to bathe in the River Yi with the breeze blowing on the Rain
Altar, and afterwards return home chanting songs.
The Master signed and said, “I am with Dian!”
Three of the disciples departed but Zeng Xi remained. He said, “What did you think of the
disciples’ sayings?”
The Master said, “Simply that you each spoke of your aims.”
Zeng Xi said, “Why did you smile at Zilu, Master?”
“He would govern a state with ritual though his words are not deferential. That is why I
smiled.”
“Is Ran Qiu not up to governing a state?”
“How could a territory of sixty to seventy or fifty to sixty li not be a state?”
“Is Gongxi Hua not up to governing a state?”
“If ancestral temples and diplomatic gatherings are not affairs of feudal states then what are
they? If Chi is only a minor assistant, who could be a major assistant?”
子路、曾皙、冉有、公西華侍坐。子曰：以吾一日長乎爾，毋吾以也。居則曰：不吾知
也！如或知爾，則何以哉？
子路率爾而對曰：千乘之國，攝乎大國之間，加之以師旅，因之以饑饉；由也為之，
比及三年，可使有勇，且知方也。
夫子哂之。求！爾何如？對曰：方六七十，如五六十，求也為之，比及三年，可使足
民。如其禮樂，以俟君子。
赤！爾何如？對曰：非曰能之，願學焉。宗廟之事，如會同，端章甫，願為小相焉。
點！爾何如？鼓瑟希，鏗爾，舍瑟而作，對曰：異乎三子者之撰。
子曰：何傷乎？亦各言其志也。曰：莫春者，春服既成，冠者五六人，童子六七人，
浴乎沂，風乎舞雩，詠而歸。
夫子喟然歎曰：吾與點也！三子者出，曾皙後。
曾皙曰：夫三子者之言何如？子曰：亦各言其志也已矣。曰：夫子何哂由也？曰：為
國以禮，其言不讓，是故哂之。
唯求則非邦也與？安見方六七十如五六十而非邦也者？
唯赤則非邦也與？宗廟會同，非諸侯而何？赤也為之小，孰能為之大？ 640
640.Lunyu translations adapted from Lau 2000.
390
• Appendix C: Lunyu 5/26 & 11/26 parallels
Hanshi
waizhuan 7
Kongzi’s travels took him to the summit of Mt. Jing, with Zilu, Zigong, and
Yan Yuan as his companions. Kongzi said, “A noble man who climbs up high
(SBCK 7/15b–16a; always gives voice to his feelings. Disciples, speak of your desires whatever
KZJY 9.11/p. 170) they may be, and I shall expound on them.
孔子遊於景山之上，子路、子貢、顔淵從。
孔子曰：君子登高必賦。小子願者，何言其願。丘將啟汝。...
Hanshi
waizhuan 9
(SBCK 9/7b–8b;
KZJY 9.12/pp. 170–
71)
Shuiyuan 15
(SBCK 15/8a–9b;
KZJY 9.13/p. 171)
Kongzi’s travels took him to the summit of Mt. Rong together with Zigong,
Zilu, and Yan Yuan. Kongzi sighed and said, “My disciples, each of you
speak of your aims and I shall survey them.”
孔子與子貢、子路、顔淵遊於戎山之上。
孔子喟然歎曰：二三子各言爾志，予將覽焉。...
When Kongzi was traveling in the north he went east to climb Mt. Nong with
Zilu, Zigong, and Yan Yuan as his companions. Kongzi sighed and said,
“Climbing up high and gazing below makes a man’s heart full of sorrow. My
disciples, each of you speak of your aims and I shall listen to them.
孔子北遊，東上農山，子路、子貢、顔淵從焉。
孔子喟然歎曰：登高望下，使人心悲。二三子者，各言爾志，
丘將聽之。...
391
• Appendix D: Lunyu 2/17 and its parallels
Lunyu 2/17
Xunzi 27(SBCK 12a–13a) Hanshi waizhuan 3
Shuiyuan (SBCK 17/16b–17a;
(SBCK 3/20a; KZJY 3.17/pp. KZJY 3.18/p. 33)
22–23)
Zilu came to see Kongzi
in full dress.
Kongzi said, “Zilu,
why are you dressed so
fancily? When the
Yangzi first emerges at
Mt. Min its spring can
fill a goblet. By the time
it reaches the fords, you
cannot cross unless you
line up boats on a windless day. Is it not because
the lower river has so
much water? Now you
are dressed in full finery
and you are clearly full
of yourself. Who in this
world would dare to remonstrate with you?”
Zilu quickly left and
changed his clothes before coming back in and
bowing humbly.
According to a tradition,
Zilu went to see Kongzi in
full finery.
Kongzi said, “Zilu, why
are you dressed so fancily?
When the Yangzi first
emerges at Mt. Min it cannot even fill a goblet. By
the time it reaches the
fords, you cannot cross
unless you line up boats
on a windless day. Is it not
because of the magnitude
of the many rivers? Now
you are dressed in full finery and you are clearly full
of yourself. Who in this
world could improve
you?”
Zilu quickly left and
changed his clothes before
coming back in and bowing humbly.
Zilu came to see Kongzi in
full dress.
Kongzi said, “Zilu, why are
you dressed so fancily?
When the Yangzi first
emerges at Mt. Min it is large
enough to fill a goblet. By the
time it reaches the fords, you
cannot cross unless you line
up the boats on a windless
day. Is it not because the lower river is fed by so many
rivers? Now you are dressed
in full finery and you are
clearly full of yourself. Who
in this world would dare to
improve you?”
Zilu quickly left and
changed his clothes before
coming back in and bowing
humbly.
子路盛服而見孔子，孔
子曰：由，是裾裾何
也？昔者江出於岷山，
其始出也，其源可以濫
觴，及其至江之津也，
不放舟，不避風，則不
可涉也。非維下流水多
邪？今女衣服既盛，顏
色充盈，天下且孰肯諫
女矣！子路趨而出，改
服而入，蓋猶若也。
傳曰：子路盛服以見孔
子。孔子曰：由、疏疏
者何也？昔者、江於
汶，其始出也，不足以
濫觴；及其至乎江之津
也，不方舟，不避風，
不可渡也，非其眾川之
多歟！今汝衣服其盛，
顏色充滿，天下有誰加
汝哉！子路趨出，改服
而入，蓋揖如也。
子路盛服而見孔子，孔子
曰：由，是襜襜者何也？
昔者江水出於岷山，其始
也，大足以濫觴，及至江
之津也，不方舟，不避
風，不可渡也，非唯下流
衆川之多乎？今若衣服甚
盛，顔色充盈，天下誰肯
加若者哉！子路趨而出，
改服而入，蓋自如也。
392
......Lunyu 2/17 .....................Xunzi 27 .........Hanshi waizhuan 3 ..............................Shuiyuan
The Master said, Kongzi said, “Zilu,
“Zilu, I will
take note, because I will
teach you about tell you: those who
knowing.”
show off their speech
are grandiloquent, and
those who show off their
conduct are arrogant.
Those who make a show
of their knowledge and
abilities are petty men.
Thus the noble man
Taking knowing takes knowing as
as knowing and knowing and not
not knowing as knowing as not knownot knowing— ing—this is the essence
this is knowing. of speech.
子曰：由！
誨女知之乎！
知之為知之，
不知為不知，
是知也。
Kongzi said, “Zilu, take
note, because I will tell
you: those who are circumspect about their speech are
not grandiloquent, and
those who are circumspect
about their conduct are not
arrogant. Those who make
a show of their knowledge
and strengths are petty men.
Thus the noble man
takes knowing as knowing and not knowing as
not knowing—this is the
essence of speech.
He takes being able as
He takes being able being able and being unable
as being able and being as being unable—this is the
essence of good conduct.
unable as being unable—this is the essence Grasping the essence of
speech is wisdom; grasping
of good conduct.
the essence of good conduct
Grasping the essence
is humaneness. If you are
of speech is wisdom;
grasping the essence of both wise and humane, then
how could you be
good conduct is huimproved?
maneness. If you are
The Odes say, ‘Tang was
both wise and humane,
then how could you be not born late / his sageliness
and respect daily
improved?”
increased.’”
孔子曰：由志之！
孔子曰：由志之，
吾語汝：奮於言者
華，奮於行者伐，色 吾語女；夫慎於言者不
譁，慎於行者不伐。色知
知而有能者，小人
也。故君子知之曰知 而有長者、小人也。故君
子知之為知之，不知為不
之，不知曰不知，
言之要也；能之曰能 知，言之要也；能之為能
之，不能曰不能，行 之，不能為不能，行之要
之至也。言要則知， 也。言要則知，行要則
仁，既知且仁，又何加
行至則仁；既仁且
哉！詩曰：湯降不遲，聖
知，夫惡有不足矣
敬日躋。
哉！
393
Kongzi said, “Zilu, take note,
because I will tell you: those
who show off their speech are
grandiloquent, and those who
show off their conduct are arrogant. Those who make a
show of their knowledge and
abilities are petty men.
Thus the noble man takes
knowing as knowing and
not knowing as not knowing—this is the essence of
speech.
He takes being able as being able and being unable as
being unable—this is the
essence of good conduct.
Grasping the essence of
speech is wisdom; grasping the
essence of good conduct is humaneness. If you are both wise
and humane, then how could
you be improved?
Zilu, the Odes say, ‘Tang
was not born late / his sageliness and respect daily increased.’ This is what I mean.”
孔子曰：由記之，
吾語若：賁於言者，華也，
奮於行者，伐也。夫色智而
有能者，小人也。故君子知
之為知之，不知為不知，
言之要也；能之為能，不能
為不能，行之至也。言要則
知，行要則仁；既知且仁，
夫有何加矣哉？由，詩曰：
湯降不遲，聖教日躋。此之
謂也。
• Appendix E: Second-order knowledge in the Laozi, Lüshi chunqiu, Zhuangzi, Huainanzi, and
Wenzi
Laozi 71 (SBCK 下/ Knowing that you do not know is highest; not knowing whether you know
17a)
is a fault. Thus a sage has no faults because he takes his faults as faults and
thereby has no faults.
知不知上，不知知，病。是以聖人不病，以其病病。是以不病。
Lüshi chunqiu 5/3, All men use their understanding to know, yet they do not know how they
“Chi yue 侈樂”
know. Knowing how one knows is called “knowing the Way”; not knowing
(SBCK 5/5a)
how one knows is called “throwing away treasure.”
人莫不以其知知，而不知其所以知。知其所以知之謂知道，不知其所
以知之謂棄寶。
Lüshi chunqiu
13/5, “Jin ting 謹
聽” (SBCK 13/10a)
The highest is knowing, the next best is knowing that you do not know.
When you do not know you ask, when you are unable you learn.
太上知之，其次知其不知。不知則問，不能則學。
Lüshi chunqiu
Knowing that one does not know is highest. The problem with those who
25/2, “Bie lei 別類 go too far is that they do not know yet think that they do know.
” (SBCK 25/3a)
知不知上矣。過者之患，不知而自以為知。
Zhuangzi 2, “Qi
wu lun 齊物論”
(SBCK 1/38a)
Nie Que asked Wang Ni, “...Do you know what you do not know?” “How
can I know that?”... “How do I know that when I say I know I really don’t
know? How do I know that when I say I don’t know I really know?”
齧缺問乎王倪曰：...子知子之所不知邪？曰：吾惡乎知之！...庸詎知
吾所謂知之非不知邪？庸詎知吾所謂不知之非知邪？
Zhuangzi 22, “Zhi Great Purity looked upward and sighed, saying, “In that case, is not knowbei you 知北遊” ing knowing? Is knowing not knowing? Who knows when not knowing is
(SBCK 7/52b)
knowing?”
泰清中而歎曰：弗知乃知乎！知乃不知乎！孰知不知之知？641
Huainanzi 12,
“Dao ying 道應”
(SBCK 12/1b)
Great Purity looked upward and sighed, saying, “In that case, is not knowing knowing? Is knowing not knowing? Who knows whether knowing is
not knowing or not knowing is knowing?”
太清仰而歎曰：然則不知乃知邪？知乃不知邪？孰知知之為弗知，弗
知之為知邪？
641.Reading zhong 中 as yang 仰 in keeping with Huainanzi version below.
394
Huainanzi 12,
“Dao ying 道應”
Thus Laozi said, “To know that you do not know is highest. Not knowing
that you know is a fault.”
(SBCK 12/14b)
故老子曰：知而不知，尚矣；不知而知，病也！
Wenzi 7, “Wei
Wenzi said, “...[if] knowing is not knowing and not knowing is knowing,
ming 微明” (Wenzi [then] who knows whether knowing is not knowing or now knowing is
shuyi 7.304)
knowing?”
文子曰：...知之乃不知，不知乃知之，孰知知之為不知，不知之為知
乎。
395
• Appendix F: Lunyu 14/34 parallels in Western Han sources
★
marks Laozi 63 parallels,
marks Lunyu 14/34 parallels only
Liji, “Biao ji 表記” The Master said, “Someone who repays wrong with virtue possesses a self(SBCK 17/2a)
expanding humaneness; someone who repays virtue with wrong deserves
punishment.”
子曰：以德報怨，則寬身之仁也；以怨報德，則刑戮之民也。
Liji, “Biao ji 表記” The Master said, “When you repay virtue with virtue, there is something to
(SBCK 17/1b)
exhort the people; when you repay wrong with wrong, there is something
to punish the people.”
子曰：以德報德，則民有所勸；以怨報怨，則民有所懲。
Xinshu, “Tui rang
退讓” (Xinshu
jiaozhu 7.284)642
There is a saying, “Turn losses into successes and follow misfortune with
fortune.” Laozi said, “Repay wrong with virtue.”
語曰：轉敗而為功，因禍而為福。老子曰：報怨以德。
Hanshu 58.2627, a [Emperor Wu] issued an edict that said: “We have heard that one should
112 BCE edict is- ‘repay virtue with virtue and repay wrong with rectitude.’”
sued by Emperor
Wu 武帝
下詔曰：朕聞報德以德，報怨以直。...
Xinxu 4, “Za shi 雜 There is a saying, “Turn losses into successes and follow misfortune with
事” (SBCK 4/7a)
fortune.” Laozi said, “Repay wrong with virtue.”
語曰：轉敗而為功，因禍而為福。老子曰：報怨以德。
Shuiyuan 13,
Kongzi said, “A sage turns misfortune into fortune and repays wrong with
“Quan mou 權謀” virtue.”
(SBCK 13/14a)
孔子曰：聖人轉禍為福，報怨以德。643
642.This line is missing in the Sibu congkan edition.
643.A few of these Lunyu 14/34 parallels deserve special mention. Notice the confusion in the Liji and Shuiyuan
parallels over the ultimate source of the Laozi saying. The “Biao ji” chapter of the Liji attributes both the Lunyu
14/34 version (“repay virtue with virtue”) and the Laozi version (“repay wrong with virtue”) to the Master without
giving any indication that either saying originated in an earlier source. The Laozi saying’s attribution to Laozi in the
Xinxu but to Kongzi in the Shuiyuan, both of which are attributed to Liu Xiang 劉向 in the late Western Han, is
equally curious. Perhaps Liu Xiang simply did not care enough about his sources to standardize these sayings’
attribution, or perhaps we should abandon the assumption that a single person was responsible for the compilation of
these texts. Whatever the explanation, the process whereby widely circulating sayings became stamped with Kongzi
yue regardless of any earlier associations seems not to have ended with the Lunyu.
396
• Appendix G: Lunyu 9/26 parallels
Parallels to the first part of Lunyu 9/26
Sunzi bingfa 7, The three armies can be robbed of their spirit, and a general can be robbed of
“Jun zheng 軍爭 his heart.
” (SBCK 7/23a)
三軍可奪氣，將軍可奪心。
Zuozhuan Xuan
12 (SBCK 11/6a)
A military motto says, “To anticipate the other, rob him of his heart.”
軍志曰：先人有奪人之心。
Zuozhuan Wen 7 Xuanzi said, “...To anticipate the other and rob him of his heart is a fine
(SBCK 8/12a)
strategem for a military man.”
宣子曰：...先人有奪人之心，軍之善謀也。
Zuozhuan Zhao
21 (SBCK 24/14b)
A military motto has it, “To anticipate another, rob him of his heart; to follow another, wait for his decline.”
軍志有之：先人有奪人之心，後人有待其衰。
Shiji 5.192–93, Inner Scribe Liao said, “The Rong king lives in a faraway land and has never
“Qin benji 秦本 heard the sounds of the central states. My lord should try presenting him
紀”
with dancing girls and musicians and thereby rob him of his will.”
內史廖曰：戎王處辟匿，未聞中國之聲。君試遺其女樂，以奪其志。
Wei Liaozi 4,
One who is adept at using an army is able to rob others without being robbed
“Zhan wei 戰威” by others. That which is robbed is the heart’s trigger, that which is com(Wei Liaozi jiaozhu manded is all the hearts in unison.
4.17)
善用兵者，能奪人而不奪於人。奪者心之機也，令者一眾心也。
Parallels to the second part of Lunyu 9/26
Liji, “Zi yi 緇衣” The Master said, “Speak about [real] things and act in accordance with
(SBCK 17/14a)
standards, thus your will cannot be robbed when you are alive and your
name cannot be robbed when you are dead.”
子曰：言有物而行有格也，是以生則不可奪志，死則不可奪名。
397
Liji, “Ru xing 儒 The Ru dwell with the men of today but investigate the ancients. They prac行” (SBCK 19/5b) tice [the Way] in the present age and later generations will take them as
models. If they do not meet with the appropriate age, those above do not support them and those below do not advance them, and factions of slandering
sycophants will endanger them. Although their persons are endangered, their
will cannot be taken from them.
儒有今人與居，古人與稽；今世行之，後世以為楷。適弗逢世，上弗
援，下弗推，讒諂之民，有比黨而危之者，身可危也，而志不可奪
也。
398
• Appendix H: Lunyu 14/35 parallels
Mao Shi #65, “Shu I go on my way, bowed down / By the cares that shake my heart. / Those
li 黍離” (SBCK 4/1b) who know me / Say, “It is because his heart is so sad.” / Those who do not
know me / Say, “What is he looking for?” Oh, azure Heaven far away, /
What sort of men can they be?644
行邁靡靡，中心搖搖。知我者，謂我心憂。不知我者，謂我何求。悠
悠蒼天，此何人哉。
Mao Shi #109,
It is my heart’s sadness / That makes me chant and sing. / Those who do
“Yuan you tao 園 not know me / Say, “My good sir, you are impudent. / That man is perfect有桃” (SBCK 5/10a) ly right. / What is this that you are saying about him?” My heart’s sorrow,
/ Which of them knows it? / Which of them knows it? / The truth is, they
do not care.645
心之憂矣，我歌且謠。不知我者，謂我士也驕。彼人是哉，子曰何
其。心之憂矣，其誰知之，其誰知之，蓋亦勿思。
Laozi 70 (SBCK 下/ My words are extremely easy to understand, and extremely easy to prac16b)
tice. But there is no one in the world who can understand them, and no one
who can practice them. My words have a progenitor, my doings have a
lord. It is because they lack understanding that they do not understand me.
When those who understand me are few, I am honored. This is why sages
clad themselves in rough garb and harbor their jade.
吾言甚易知，甚易行。天下莫能知，莫能行。言有宗，事有君。夫唯
無知，是以不我知。知我者希，則我者貴。是以聖人被褐懷玉。
Shanghai Museum The Master sighed and said, “Ah! No one understands me!” Ziyou said,
“Dizi wen 弟子問” “Are you speaking about enacting [your Way]?”
ms., strip ＃4
(Shanghai bowuguan 子嘆曰：烏！莫我知也夫！子游曰：有施之謂乎？...
cang Zhanguo Chu
zhushu vol. 5, p. 269)
644.Translation from Waley 1996, p. 56.
645.Translation from Waley 1996, p. 86.
399
Mawangdui “Laozi My words are extremely easy to understand, and extremely easy to pracjia ben 老子甲本” tice. But no one else can understand them, and no one else can practice
ms. (Mawangdui Han them. My words have a lord, my doings have a progenitor. It is because
mu boshu, p. 6)
they lack understanding that they do not understand me. When those who
understand me are few, I am honored. This is why sages clad themselves
in rough garb and harbor their jade.
吾言甚易知也，甚易行也；而人莫之能知也，而莫之能行也。言有
君，事有宗。夫唯⎔知也，是以不我知。知我者希，則我貴矣。是以
聖人被褐而褱（懷）玉。
Chuci, “Li sao 離
騷” (SBCK 1/49a)
Coda: I am through! There is no one in my state who understands me, so
why should I cleave to the city of my birth? Since there is no one worthy
to work with in making good government, I shall go and join Peng Xian in
the place where he abides.646
亂曰：已矣哉，國無人莫我知兮，又何懷乎故都？既莫足與為美政
兮，吾將從彭咸之所居。
Chuci, “Huai sha The courtier crowd are low and vulgar fellows; / They cannot understand
懷沙” (Shiji 84.2487) the things I prize.
夫黨人之鄙妒兮，羌不知吾所臧。
Art and nature perfected lay within me hidden; / But the crowd did not
know of the rare gifts that were mine. / Unused materials I had in rich
store; / Yet no one knew the things that I possessed.
文質疏內兮，眾不知吾之異采；材樸委積兮，莫知余之所有。
Coda: The mighty waters of the Yuan and Xiang with surging swell go
rolling on their way; / The road is long, through places dark and drear, a
way far and forlorn. / The nature I cherish in my bosom, the feelings I embrace, there is no one in the world who understands me. The heart of man
cannot be told...Since there is no one in the world who understands me /
the heart of man cannot be told.647
亂曰：浩浩沅、湘兮，分流汨兮。脩路幽拂兮，道遠忽兮。曾吟恆悲
兮，永歎慨兮。世既莫吾知兮，人心不可謂兮。...世溷不吾知，心不
可謂兮。
Shiji 84.2494, Jia
Yi’s 賈誼 “Diao
Qu Yuan fu 弔屈
原賦”
Coda: I am through! There is no one in my state who understands me.
Alone and bottled up with despair, who can I tell?
訊曰：已矣，國其莫我知，獨堙鬱兮其誰語？
646.Translation adapted from Hawkes 1985, p. 78.
647.Translation adapted from Hawkes 1985, pp. 169–172.
400
• Appendix I: Lunyu 14/21 and the Zuozhuan
Lunyu 14/21
Zuozhuan Ai 15 (SBCK 30/4a; KZJY p. 425)
On jiawu, Chen Huan of Qi assassinated his lord
Ren at Shuzhou. Kong Qiu purified himself for
three days and begged [Duke Ai] to attack Qi. After
the third entreaty the Duke said, “Lu is much weakChen Chengzi assassinated Duke Jian. Kongzi per- er than Qi. If you would have us attack them then
formed ablutions before going to court to inform
would would happen?” He responded, “Chen Huan
Duke Ai, “Chen Heng assassinated his lord. I beg assassinated his lord so half of the people do not
you to punish him.”
support him. We can defeat him with Lu’s strength
and half of Qi’s.
陳成子弒簡公。孔子沐浴而朝，告於哀公曰：
陳恆弒其君，請討之。
甲午，齊陳恆弒其君壬于舒州。孔丘三日齊而
請伐齊，三，公曰：魯為齊弱久矣。子之伐
之，將若之何。對曰：陳恆弒其君，民之不與
者半，以魯之眾，加齊之半，可克也。
The Duke said, “Tell the three ministers [i.e., the
The Duke said, “Inform the Jisuns.”
heads of the three great ministerial lineages in Lu].”
公曰：告夫三子。
公曰：子告季孫。
Kongzi said, “I am ranked after the grandees so I
Kongzi declined then retired and announced to othdid not dare not to inform you. [Yet now] My Lord ers, “I am ranked after the grandees so I did not
says, ‘Tell the three sons.’” Kongzi reported this to dare not to speak.”
the three sons but his request was denied.
孔子曰：以吾從大夫之後，不敢不告也。君曰
告夫三子者！之三子，告，不可。
孔子辭，退而告人曰：吾以從大夫之後也。故
不敢不言。
Kongzi said, “I am ranked behind the grandees and
so I did not dare not to inform you.”
孔子曰：以吾從大夫之後，不敢不告也。648
648.This line also appears at Lunyu 11/8 but in a very different context.
401
• Appendix J: Lunyu 2/19 and the Huainanzi
Lunyu 2/19649
Huainanzi 16, “Shui shan 說山“ (SBCK 16/11a)
When the Jisun clan seized control of the ducal
clan Kongzi tried to persuade them by conforming
to their conduct before entering their government.
Duke Ai asked, “What can I do to make the people
submit?”
季孫氏劫公家，孔子說之。650
哀公問曰：何為則民服？
先順其所為，而後與之入政。
Kongzi responded,
孔子對曰：
Kongzi said,
曰：
“Raise the straight and set it against the crooked
And the people will submit.
Raise the crooked and set it against the straight,
And the people will not submit.
“Raising the crooked in service of the straight—
How can it not succeed?
Raising the straight in service of the crooked—
That should not be followed.”
舉直錯諸枉(*-aŋʔ)，
舉枉與直(*-əәk)，
則民服(*-əәk)；
舉枉錯諸直(*-əәk)，
則民不服(*-əәk)。
如何而不得(*-əәk)？
舉直與枉(*-aŋʔ)，
勿與遂往(*-aŋʔ)。
This is what is called sharing the same defilement
but for different ends...
此所謂同汙而異途者。
649.Also compare Lunyu 12/22: “The Master said, ‘Raising the straight against the crooked can make the crooked
straight” (子曰：「舉直錯諸枉，能使枉者直。).
650.By contrast, Sarah Queen and John Major in their translation of Huainanzi book 16 read the character 說 (here
read as shui “persuade”) ias yue (to be pleased). See Major et al. 2010, p. 650.
402
• Appendix K: Lunyu 9/28 parallels
Lüshi chunqiu Kongzi said, “...Only after the great cold has arrived and the frost and snow
14/6, “Shen
have fallen do I know that pine and cedar still flourish.”
ren 慎人”
(SBCK 17b–18a) 孔子曰：...大寒既至，霜雪既降，吾是以知松柏之茂也。
Zhuangzi 28, Kongzi said, “...Only after heaven’s cold has arrived and the frost and snow
“Rang wang have fallen do I know that pine and cedar still flourish.”
讓王” (SBCK
9/28a)
孔子曰：...天寒既至，霜雪既降，吾是以知松柏之茂也。
Xunzi 27, “Da The noble man when at rock bottom does not lose [his Way], when exhausted is
lüe 大略”
not careless, and when encountering great difficulties does not forget the words
(SBCK 19/16b– he once spoke on the mats [i.e., in less difficult times]. If the year did not grow
17a)
cold there would be no way to know the pine and cypress, and if one encountered no difficulties there would be no way to know that not a day goes by
when the noble man is not steadfast.651
君子隘窮而不失，勞倦而不苟，臨患難而不忘細席之言。歲不寒無以知松
柏，事不難無以知君子無日不在是。
Huainanzi 2, Only after the great cold has arrived and the frost and snow have fallen do I
“Chu zhen 俶 know that pine and cedar still flourish. Bearing difficulties and encountering
真” (SBCK 2/3b) dangers, with benefit and harm arrayed before him—only then can it be known
that sage does not lose the Way.
夫大寒至，霜雪降，然後知鬆柏之茂也。據難履危，利害陳於前，然後知
聖人之不失道也。
651.Here I follow Yang Liang’s suggestion and read xi 細 (slight) as xi 昔 (former). See SBCK 19/17a.
403
• Appendix L: Lunyu 9/17 parallels
Mengzi 4B/18 (SBCK Xuzi said, “Zhongni often praised water, saying, ‘Water! Water!’ What
8/6a)
did he glean from water?”...
徐子曰：仲尼亟稱於水，曰：水哉，水哉！』何取於水也。孟子
曰：「（源）〔原〕泉混混，不舍晝夜，盈科而後進，放乎四海。
有本者如是，是之取爾。苟為無本，七八月之間雨集，溝澮皆盈；
其涸也，可立而待也。故聲聞過情，君子恥之。
Xunzi book 28,
“You zuo 宥坐”
(SBCK 20/5b)
Kongzi gazed upon the eastward flow of the waters. Zigong asked
Kongzi, “Why when a junzi sees great waters does he always gaze at
them?” Kongzi said, “Water extends greatly to all living things yet without acting, like virtue...”
孔子觀於東流之水，子貢問於孔子曰：君子之所以見大水必觀焉
者，何也？孔子曰：夫水大遍與諸生而無為也，似德...
Shizi, “Ming tang
Kongzi said, “Great are the rivers and seas!...”
明堂” (in SBCK Qunshu zhiyao 36/14a)
孔子曰：大哉河海乎！...
Fuyang “Rujia zhe
yan 儒家者言”
Kongzi looked upon the River and sighed.
(Fuyang Hanjian
“Zhouyi” yanjiu, p.
157)
孔子臨河而歎。
Da Dai Liji 64,
“Quan xue 勸學”
Zigong said, “Why when a junzi sees a great river does he always gaze at
it?” Kongzi said, “Water is what a junzi likens to virtue.”
(SBCK 7/8b)
子貢曰：君子見大川必觀，何也？孔子曰：夫水者，君子比德焉。...
Shuiyuan 17, “Za
yan 雜言” (SBCK
Zigong said, “Why when a junzi sees a great river does he always gaze at
it?” Kongzi said, “Water is what a junzi likens to virtue.”
17/22a)
子貢問曰：君子見大水必觀焉，何也？孔子曰：夫水者，君子比德
焉。...
Kongzi jiayu 9,
“San shu 三恕”
(SBCK 2/13b)
Kongzi gazed upon the eastward flow of the waters. Zigong asked, “Why
is it that a junzi when he sees great waters always gazes at them?” Kongzi
said, “It is because they do not rest, and extend to all living things yet
without acting...”
孔子觀於東流之水，子貢問曰：君子所〔以〕見大水必觀焉，何
也？孔子（對）曰：以其不息，且遍與諸生而不為也。
404
• Appendix M: Lunyu 9/17 and the Chuci
Chuci “Jiu ge 九 As with trembling heart I gaze on the distance,
歌”, “Xiang furen Contemplating the swiftly moving waters.
湘夫人” (SBCK
...
2/13a)
Oh, rarely, rarely the time is given!
I wish I could play here a little longer.652
荒忽兮遠望，觀流水兮潺湲。...時不可兮驟得，聊逍遙兮容與。
Chuci “Jiu zhang The way is so far, that each day I remember less;
九章”, “Chou si And I wish to make my plaint, but I cannot, for none will hear.
抽思” (SBCK
I gaze on the northern hills and my tears come falling,
4/21b)
Look down on the flowing waters and heave a dolorous sigh.653
道卓遠而日忘兮，願自申而不得。望北山而流涕兮，臨流水而太息。
Chuci, “Xi shi 惜 Oppressed by each day’s new signs of age and decay,
誓” (SBCK 11/1b) By the swift, irreversible passage of the years,
I climbed the blue heaven, mounted up on high,
And, passing down over a myriad peaks, farther and farther I flew.
I contemplated the meanders of great rivers,
Came where the four seas drenched me with their spray.654
惜余年老而日衰兮，歲忽忽而不反。登蒼天而高舉兮，歷眾山而日
遠。觀江河之紆曲兮，離四海之霑濡。
Chuci “Qi jian 七 Better to throw myself into the river’s waters
諫”, “Yuan shi
And set my spirit hurrying across its swift currents (shi 逝).
怨世” (SBCK
I would rather become mud in the bed of the sea or the river
13/13b–14a)
Than look any longer on this unclean age.655
願自沈於江流兮，絕橫流而徑逝。寧為江海之泥塗兮，安能久見此濁
世？
652.Translation adapted from Hawkes, p. 108.
653.Translation adapted from Hawkes, p. 168.
654.Translation adapted from Hawkes, p. 240.
655.Translation adapted from Hawkes, pp. 251–252.
405
• Appendix N: Lunyu 12/1 and the Zuozhuan
Zuozhuan Zhao
12 (SBCK 22/15b;
KZJY p. 418)
Zhongni said, “There is an ancient maxim: ‘Subduing oneself and reviving
ritual is humaneness.’ How true! If King Ling of Chu had been like this how
could he have been shamed at Ganxi?”
仲尼曰：古也有志：克己復禮，仁也。信善哉！楚靈王若能如是，豈其
辱於乾谿。
Lunyu 12/1
Yan Yuan asked about humaneness.
The Master said, “Subduing oneself and reviving ritual is humaneness. If in
one day one can subdue himself and revive ritual then the whole world will
return to humaneness. One becomes humane from oneself—how could it be
from another?”
Yan Yuan said, “May I ask about the particulars?”
The Master said, “If it is not ritually proper then do not look at it; if it is not
ritually proper then do not listen to it; if it is not ritually proper then do not
speak it; if it is not ritually proper then do not act on it.”
Yan Yuan said, “Even though I am not intelligent I beg to put this saying into
practice.”
顏淵問仁。
子曰：克己復禮為仁。 一日克己復禮，天下歸仁焉。為仁由己，而由
人乎哉？顏淵曰：請問其目。
子曰：非禮勿視，非禮勿聽，非禮勿言，非禮勿動。
顏淵曰：回雖不敏，請事斯語矣。
406
• Appendix O: Lunyu 12/2 and the Zuozhuan
Zuozhuan Xi Your subject [Ji Que] has heard, “When away from home act as though you are
33 (SBCK 7/15a) a guest and undertake your duties as though you are sacrificing. This is the rule
for humaneness.”
臣[冀缺]聞之：出門如賓，承事如祭，仁之則也。
Lunyu 12/2
Zhonggong asked about humaneness. The Master said, “When away from
home act as though you are seeing an important guest; when directing the people act as though you are conducting a great sacrifice.” Do not impose on others what you yourself do not want. [Thus] you will incur no resentment in your
state or in your home.
仲弓問仁。子曰：出門如見大賓，使民如承大祭。己所不欲，勿施於
人。在邦無怨，在家無怨。
407
• Appendix P: Lunyu 7/1 parallels
Lunyu 7.1
The Master said: “Insofar as I transmit without originating and trust in
and love the ancient, I would liken myself to Old Peng.”
子曰：述而不作，信而好古，竊比於我老彭。
1. Mozi 39, “Fei ru xia 非儒 [The Ru] also say: “A noble man follows without originating.”
下” (SBCK 9/18a)
[儒者]又曰：君子循而不作 。
2. Mozi 46, “Geng zhu 耕柱 Gong Mengzi said: “A noble man does not originate; he only transmits.”
” (SBCK 11/16b)
公孟子曰：君子不作，術而已。
3. Xunzi 荀子 22, “Zheng
ming 正名” (SBCK 16/3b)
If a true king was to arise, he would certainly follow the old names in
some instances and originate new names in others.
若有王者起，必將有循於舊名，有作於新名。
4. Huainanzi 13, “Fan lun 氾 A great man originates and a follower follows.
論” (SBCK 13/5b)
大人作而弟子循。
5. Liji, “Zhong yong 中庸”
(SBCK 16/5a)
6. Liji, “Yueji 樂記” (SBCK
11/9a)
7. Shiji 史記 127.3218
The Master said, “...a father originates, a son transmits.”
子曰：...父作之，子述之。
Thus those who know the nature of Ritual and Music can create; those
who recognize the patterns of Ritual and Music can transmit. Originators
are called sagely; transmitters are called enlightened. Being enlightened
or sagely refers to transmitting or creating.656
故知禮樂之情者能作，識禮樂之文者能述。作者之謂聖，述者之謂
明。明聖者，述作之謂也。
Sima Jizhu said: “...transmitting without originating is the principle of the
noble man.
司馬季主曰：...述而不作，君子義也...
8. Shiji 130.3299
The Grand Scribe says: “...There is no crime greater than abandoning the
(“Taishigong zixu 太史公自 flourishing virtue of the sagely and enlightened without recording it, or
序”)
destroying the heritage of meritorious ministers, hereditary houses, and
accomplished grandees without transmitting it, or letting the words of
those who came before fall away. What I refer to as ‘transmitting’ past affairs and organizing the traditions of each generation is not an instance of
‘originating.’ You are mistaken, sir, if you compare this work to the
Spring and Autumn Annals.”
太史公曰： ...廢明聖盛德不載，滅功臣世家賢大夫之業不述，墮先
人所言，罪莫大焉。余所謂述故事，整齊其世傳，非所謂作也，而
君比之於春秋，謬矣。
656.Translation after Cook 1995, p. 46.
408
9. Yantie lun 23, “Zun dao
遵道” (SBCK 5/11b)
The Man of Culture and Learning said, “The noble man is well informed,
he transmits without originating, he is sagely and penetrating with great
plans, and he is wise with few affairs.”
文學曰：君子多聞闕疑，述而不作，聖達而謀大，叡智而事寡。657
10. Yantie lun 10, “Ci fu 刺
復”) (SBCK 2/10b)
The Man of Culture and Learning said, “Raising up the square and compass to let others understand what is appropriate and blowing the pitchpipes to let others understand variation is superior; following without
originating and awaiting others is second-best.
文學曰：...夫舉規矩而知宜，吹律而知變，上也；因循而不作，以
俟其人，次也。
11. Fayan, “Wen ming 問神 Some say: “[Given that one should] ‘transmit without originating,’ why
” (SBCK 5/3b)
did you originate the Canon of Supreme Mystery?
或曰：述而不作，玄何以作？
12. Hanshu 22.1071, “Liyue Restoring the old culture, doing away with [the melodies of] Zheng and
zhi 禮樂志”
drawing close to the elegant, “transmitting without originating, trusting in
and loving the ancient.”
修起舊文，放鄭近雅，述而不作，信而好古。
13. Hanshu 24.1142, “Shihuo zhi 食貨志”, advice to
Emperor Ai 哀 by Shi Dan
師丹.
When conducting government the noble man honors following and treats
reform and originating as weighty affairs.
蓋君子為政，貴因循而重改作。
14. Hanshu 88.3589, “Rulin And so [Kongzi] said: “Transmitting without originating, trusting in and
liezhuan 儒林列傳”
loving the ancient.”
故[孔子]曰：述而不作，信而好古。
15. Hanshu 97.3980, “Waiqi The Way of the noble man is to delight in following and to treat reform
liezhuan 外戚列傳”, an im- and originating as weighty affairs.
perial “response” (bao 報)
君子之道，樂因循而重改作。
drafted by Liu Xiang 劉向
and Gu Yong 谷永 for Emperor Ai
16. Lunheng 81, “Zheng
shuo 正說” (SBCK 28/4a)
The sages make ‘classics,’ accomplished men make ‘writings.’
聖人作經，賢者作書。
17, Lunheng 84, “Dui zuo 對 Some say: “Sages originate, and accomplished men transmit. It is wrong
作篇” (SBCK 29/8b)
for a (merely) accomplished man to create. The Balanced Discourses and
Government Affairs could be called ‘creations.’” In reply I say: “They are
neither ‘original works’ nor ‘transmissions.’ They are ‘discussions.’”
或曰：聖人作，賢者述，以賢而作者，非也。論衡、政務，可謂作
者。曰：非作也，亦非述也，論也。
18. Hou Hanshu 44.1500–
01, a memorial by Xu Fang
徐防 submitted ca. 103 CE
Kongzi said, “Transmit without originating.”
孔子稱：述而不作。
657.Lunyu 2/18: 多聞闕疑
409
19. Zhao Qi’s 趙岐 (d. 201
CE) commentary to Mengzi
4A (SBCK 7/1a)
In this way one must rely on the square and compass to make things
square or round, just like the Lunyu’s “transmitting without originating,
trusting in and loving the ancient.”
然必須規矩，乃成方圓，猶論語述而不作，信而好古。
410
• Appendix Q: “Erring” (guo 過) and “reforming” (gai 改) in the Lunyu and elsewhere
Lunyu 1/8
The Master said, “Taking loyalty and trustworthiness as one’s master; not becoming friends with someone who is inferior to oneself; erring and not fearing reform.”
子曰：主忠信。無友不如己者。過則勿憚改。
Lunyu 7/3
The Master said, “Having virtue without refining it; learning without practicing it; hearing what is right without moving to do it; having defects without
reforming them—these are my worries.”
子曰：德之不脩，學之不講，聞義不能徙，不善不能改，是吾憂也。
Lunyu 7/22
The Master said, “When walking in a trio there is always one who can be my
teacher. I select their good points and follow them, and from their bad points
I reform myself.”
子曰：三人行，必有我師焉，擇其善者而從之，其不善者而改之。
Lunyu 9/25
The Master said, “Taking loyalty and trustworthiness as one’s master; not becoming friends with someone who is inferior to oneself; erring and not fearing reform.”
子曰：主忠信，毋友不如己者，過則勿憚改。
Lunyu 15/30
The Master said, “To err without reforming—this is an error.”
子曰：過而不改，是謂過矣。
Lunyu 19/21
Zigong said, “The junzi’s errors are like an eclipse of the sun or moon: when
he errs, all people see it, and when he reforms himself, all people look up to
him.”
子貢曰：君子之過也，如日月之食焉：過也，人皆見之；更也，人皆仰
之。
Zuozhuan Xuan [Duke Ling of Jin] said, “When I understand where I have erred, I will re宣 2 (SBCK 10/3b) form myself.” Shiji kowtowed and responded, “What man hasn’t erred? To
err and be able to reform—there is nothing finer than this....”
[晉靈公]曰：吾知所過矣，將改之。[士季]稽首而對曰：人誰無過，過
而能改，善莫大焉。
Mengzi 2B/9
(SBCK 4/11b)
Mengzi said, “And when ancient junzi erred they reformed themselves; when
junzi today err they follow [the error]...”
孟子曰：...且古之君子，過則改之；今之君子，過則順之。...
411
Mengzi 6B/19
(SBCK 12/17a)
Mengzi said, “People constantly err, but afterwards they are able to reform
themselves.”
孟子曰：......人恆過，然後能改。
Zhouyi #42, “Yi The Xiang commentary: a junzi sees the good and moves toward it, and when
益” (SBCK 4/14a) he errs he reforms.
象曰：...君子以見善則遷，有過則改。
Guoyu, “Lu yu
shang 魯語上”
Wenzi learned of this and said, “One who errs and is able to reform is the
highest kind of person.”
(SBCK 4/16a)
文子聞之，曰：過而能改者，民之上也。
Guanzi 31, “Jun A lord who errs without reforming is said to “collapse.”
chen xia 君臣下
” (SBCK 11/2b)
君有過而不改，謂之倒。
Chuci, “Tian
wen 天問”
To realize errors and reform them—what do we say of this?
(SBCK 3/34a)
悟過改更，我又何言？
Guliang Xi 22
To err without reforming—this is called an “error.”
(SBCK 5/18a)
過而不改，又之是謂之過。
Da Dai Liji 49, Zengzi said, “To err without being able to reform is laxity.”
“Zengzi lishi 曾
子立事” (SBCK 曾子曰 ：...過而不能改，倦也。
4/4b–5a)
Da Dai Liji 66, To err is to miss. The human condition is such that there is no one who does
“Sheng de 盛德 not err. But erring and reforming it—this is not an error.
” (SBCK 8/9b)
過，失也。人情莫不有過，過而改之，是不過也。
Shuiyuan 1,
The junzi said, “To err and reform it—this is not to err.”
“Jun dao 君道”
(SBCK 1/17a)
君子曰：夫過而改之，是猶不過。
412
• Appendix R: The junzi 君子/xiaoren 小人 dichotomy in the Lunyu
Lunyu 2/14
子曰：君子周而不比，小人比而不周。
Lunyu 4/11
子曰：君子懷德，小人懷土；君子懷刑，小人懷惠。
Lunyu 4/16
子曰：君子喻於義，小人喻於利。
Lunyu 6/13
子謂子夏曰：女為君子儒。無為小人儒。
Lunyu 7/37
子曰：君子坦蕩蕩，小人長戚戚。
Lunyu 12/16
子曰：君子成人之美，不成人之惡。小人反是。
Lunyu 12/19
...孔子對曰：...君子之德風，小人之德草。草上之風，必偃。
Lunyu 13/23
子曰：君子和而不同，小人同而不和。
Lunyu 13/25
子曰：君子易事而難說也。說之不以道，不說也；及其使人也，器之。
小人難事而易說也。說之雖不以道，說也；及其使人也，求備焉。
Lunyu 13/26
子曰：君子泰而不驕，小人驕而不泰。
Lunyu 14/6
子曰：君子而不仁者有矣夫，未有小人而仁者也。
Lunyu 14/23
子曰：君子上達，小人下達。
Lunyu 15/2
...子曰：君子固窮，小人窮斯濫矣。
Lunyu 15/21
子曰：君子求諸己，小人求諸人。
Lunyu 15/34
子曰：君子不可小知而可大受也，小人不可大受而可小知也。
Lunyu 16/8
子曰：君子有三畏：畏天命，畏大人，畏聖人之言。小人不知天命而不
畏也，狎大人，侮聖人之言。
Lunyu 17/4
...子游對曰：昔者偃也聞諸夫子曰：君子學道則愛人，小人學道則易使
也。
Lunyu 17/23
子路曰：君子尚勇乎？子曰：君子義（以）〔之〕為上，君子有勇而無
義為亂，小人有勇而無義為盜。
413
• Appendix S: The “states with the Way”/“states without the Way” dichotomy in the Lunyu
Lunyu 5/2
The Master said of Nan Rong, “In a state with the Way he will not be cast aside
and in a state without the Way he will avoid punishment and execution.”...
子謂南容：邦有道，不廢；邦無道，免於刑戮。...
Lunyu 5/21
The Master said, “As for Ning Wuzi, in a state with the Way he will be [considered] wise but in a state without the Way he will be [considered] foolish. His wisdom can be attained but his foolishness cannot.
子曰：甯武子，邦有道，則知；邦無道，則愚。其知可及也，其愚不可及
也。
Lunyu 8/13
The Master said, “...When the world has the Way be visible but when it lacks the
Way be hidden. To be poor and lowly in a state with the Way is shameful; to be
rich and honored in a state without the Way is [also] shameful.”
子曰：...天下有道則見，無道則隱。邦有道，貧且賤焉，恥也；邦無道，富
且貴焉，恥也。
Lunyu 14/1
Xian asked about shame. The Master said, “Earn a salary in a state with the Way,
but to earn a salary in a state without the Way is shameful.”
憲問恥。子曰：邦有道、穀，邦無道、穀，恥也。
Lunyu 14/3
The Master said, “In a state with the Way [one can be] perilously high-minded in
word and deed. In a state without the Way [one can be] perilously high-minded in
deed but his words must be compliant.”658
子曰：邦有道，危言危行；邦無道，危行言孫。
Lunyu 15/7
The Master said, “How straight is Shi Yu! In a state with the Way he is like an arrow and in a state without the Way he is like an arrow. What a junzi is Qu Boyu!
In a state with the Way he serves and in a state without the Way he rolls himself
up and tucks himself away.”
子曰：直哉史魚！邦有道，如矢；邦無道，如矢。君子哉蘧伯玉！邦有
道，則仕；邦無道，則可卷而懷之。
658.Translation adapted from Lau 2000, p. 133.
414
Lunyu 16/2
Kongzi said, “In a world with the Way ritual and music and military campaigns
originate from the son of heaven; in a world without the Way, ritual and music
and military campaigns originate from the vassal lords. When they originate from
the feudal lords, [their power] will likely be lost within ten generations; when
they originate from grandees, [their power] will likely be lost in five generations ;
and if their vassal subjects hold the mandate of the state then [their power] will
likely be lost in three generations. When the world has the Way then government
does not reside in the grandees; when the world has the Way, then the masses do
not squabble.
孔子曰：天下有道，則禮樂征伐自天子出；天下無道，則禮樂征伐自諸侯
出。自諸侯出，蓋十世希不失矣；自大夫出，五世希不失矣；陪臣執國
命，三世希不失矣。天下有道，則政不在大夫。天下有道，則庶人不議。
415
• Appendix T: The “states with the Way”/“states without the Way” dichotomy outside the Lunyu
1. Laozi 46 (SBCK 天下有道，卻走馬以糞；天下無道，戎馬生於郊。
下/5a)
2. Han Feizi 21,
“Yu Lao 喻老”
(SBCK 7/1a)
3. Han Feizi 36,
“Nan yi 難一”
天下有道，無急患則曰靜，遽傳不用，故曰：『卻走馬以糞。』
天下無道，攻擊不休，相守數年不已，甲冑生蟣蝨，鷰雀處帷幄，而
兵不歸，故曰：『戎馬生於郊。』
君有道，則臣盡力而姦不生；無道，則臣上塞主明而下成私。
(SBCK 15/3b)
4. Mengzi 4A/7
天下有道，小德役大德，小賢役大賢；天下無道，小役大，弱役強。
(SBCK 7/5b)
5. Mengzi 7A/42
天下有道，以道殉身；天下無道，以身殉道。
(SBCK 13/17b)
6. Zhuangzi 4,
[楚狂接輿曰；...] 天下有道，聖人成焉；天下無道，聖人生焉。
“Ren jian shi 人間
世” (SBCK 2/27b)
7. Zhuangzi 12,
“Tian di 天地”
[華封人曰：] 夫聖人鶉居而彀食，鳥行而無彰；天下有道，則與物皆
昌；天下無道，則修德就閒。
(SBCK 5/7a)
8. Guanzi 64,
“Xing shi 形勢”
故有道則民歸之，無道則民去之；故曰：道往者，其人莫來。道來
者，其人莫往。
(SBCK 20/9b)
9. Liji “Zhong
子曰：...國有道，不變塞焉，強哉矯。國無道，至死不變，強哉矯。...
yong 中庸” (SBCK 國有道，其言足以興。國無道，其默足以容。...
16/2b & 16/10b)
10. Liji “Biao ji 表 君子不以辭盡人，故天下有道，則行有枝葉。天下無道，則辭有枝
記” (SBCK 17/8a)
葉。
11. Huainanzi 10, 君，根本也；臣，枝葉也。根本不美，枝葉茂者，未之聞也。有道之
“Miu cheng 繆稱” 世，以人與國；無道之世，以國與人。
(SBCK 10/9b)
12. Huainanzi 14, 有道者，不失時與人；無道者，失於時而取人。
“Quan yan 詮言”
(SBCK 14/12b)
13. Wenzi 1,
文子曰：...有道則隱，無道則見...故有道則和，無道則苛。
“Yuan dao 原道”
(Wenzi shuyi 1.18)
14. Shiji 26.1258
天下有道，則不失紀序；無道，則正朔不行於諸侯。
15. Shiji 62.2134
國有道，即順命；無道，即衡命。
416
16. Da Dai Liji 56 曾子曰：天下有道，則君子訢然以交同；天下無道，則衡言不革。...
“Zengzi zhi yan
國有道，則突若入焉；國無道，則突若出焉，如此之謂義。
曾子制言” (SBCK
5/5a)
17. Da Dai Liji 60 國家有道，其言足以生；國家無道，其默足以容，蓋桐提伯華之行
“Wei jiangjun
也。...是故君擇臣而使之，臣擇君而事之，有道順君，無道橫命；晏
Wenzi 衛將軍文 平仲之行也。
子” (SBCK 6/9a)
18. Hanshi
君子之居也，綏如安裘，晏如覆杅。天下有道，則諸侯畏之；天下無
waizhuan 9 (SBCK 道，則庶人易之。
9/11a)
19. Chunqiu fanlu 仲尼曰：國有道，雖加刑，無刑也；國無道，雖殺之，不可勝也。
31, “Shen zhi
yang zhong yu yi
身之養重於義”
(SBCK 9/2b)
20. Wu Yue Chun- 禹曰：天下有道，民不罹辜；天下無道，罪及善人。
qiu (SBCK 6/4b)
21. Lunheng 51,
“Zhi rui 指瑞”
儒者說鳳皇騏驎為聖王來，以為鳳皇騏驎，仁聖禽也，思慮深，避害
遠，中國有道則來，無道則隱。
(SBCK 17/1a)
22. Kongzi jiayu 子曰：國無道，隱之可也；國有道，則袞冕而執玉。
9, “San shu 三恕”
(SBCK 2/16b)
23. Kongzi jiayu 孔子曰：古之士者，國有道則盡忠以輔之，無道則退身以避之。
41, “Zheng lun jie
正論解” (SBCK
9/28a)
417
• Appendix U: “Wise men”/“humane men”/“brave men” comparisons in the Lunyu
Lunyu 4/2
子曰：不仁者不可以久處約，不可以長處樂。仁者安仁，知者利仁。
Lunyu 6/23
The Master said, “The understanding delight in water, the humane delight in
mountains. The understanding act, the humane are still. The understanding are
joyful, the humane are long-lived.”
子曰：知者樂水，仁者樂山。知者動，仁者靜。知者樂，仁者壽。
Lunyu 9/29
The Master said, “A wise man is not confused, a brave man is not anxious, and a
brave man is not fearful.
子曰：知者不惑，仁者不憂，勇者不懼。
Lunyu 14/4
The Master said, “...The humane always have courage, but the courageous do not
always have humaneness.”
子曰：...仁者必有勇，勇者不必有仁。
Lunyu 14/28 The Master said, “A noble man proclaims three things, none of which are within
my abilities: a humane man is not anxious, a wise man is not confused, and a
brave man is not fearful.” Zigong said, “The Master describes himself.”
子曰：君子道者三，我無能焉：仁者不憂，知者不惑，勇者不懼。子貢
曰：夫子自道也。
418
• Appendix V: “Wise men”/“humane men”/“brave men” comparisons outside the Lunyu
Mengzi 7A/46
(SBCK 13/19a)
孟子曰：知者無不知也，當務之為急；仁者無不愛也，急親賢
之為務。
Zhuangzi 28,
湯又讓瞀光，曰：知者謀之，武者遂之，仁者居之，古之道
“Rang wang 讓王 也。
” (SBCK 9/30a)
Lüshi chunqiu
19/1 (SBCK 19/2a)
湯又讓於務光曰：智者謀之，武者遂之，仁者居之，古之道
也。
Huainanzi 2,
若夫神無所掩，心無所載，通洞條達，恬漠無事，無所凝滯，
“Chu zhen 俶真” 虛寂以待，勢利不能誘也，辯者不能說也，聲色不能淫也，美
(SBCK 2/11a)
者不能濫也，智者不能動也，勇者不能恐也，此真人之道也。
Huainanzi 9,
“Zhu shu 主術”
仁者愛其類也，智者不可惑也。
(SBCK 9/21b)
Huainanzi 14,
智者不以位為事，勇者不以位為暴，仁者不以位為患，可謂無
“Quan yan 詮言” 為矣。
(SBCK 14/6b)
Huainanzi 15,
“Bing lüe 兵略”
夫仁勇信廉，人之美才也，然勇者可誘也，仁者可奪也，信者
易欺也，廉者易謀也。
(SBCK 15/10b)
Huainanzi 18,
“Ren jian 人間”
仁者不以欲傷生，知者不以利害義。
(SBCK 18/9b–10a)
Huainanzi 20,
“Tai zu 泰族”
故仁知，人材之美者也。所謂仁者，愛人也；所謂知者，知人
也。
(SBCK 20/18b)
Shiji 43.1843,
李兌謂肥義曰：...仁者愛萬物而智者備禍於未形，不仁不智，
“Zhao shijia 趙世 何以為國？
家”
Shiji 67.2198,
“Zhongni dizi
liezhuan 仲尼弟
子列傳”
子貢曰：...夫勇者不避難，仁者不窮約，智者不失時，王者不
絕世，以立其義。
Liji, “Yue ji 樂記 強者脅弱，眾者暴寡；知者詐愚，勇者苦怯。
” (SBCK 11/7b)
Liji, “Biao ji 表記 子曰：...仁者安仁，知者利仁。畏罪者強仁。仁者右也，道者
” (SBCK 17/2a)
左也。仁者人也。道者義也。
Liji, “Sang fu si
zhi 喪服四制”
仁者可以觀其愛焉，知者可以觀其理焉，強者可以觀其志焉。
(SBCK 20/17a)
419
Guliang (SBCK
知者慮，義者行，仁者守，有此三者，然後可以出會。
1/3b)
Guliang (SBCK
知者慮，義者行，仁者守，有此三者備，然後可以會矣。
2/13a)
Wenzi 3, “Jiu shou 若夫神無所掩，心無所載，通洞條達，澹然無事，勢利不能
九守” (Wenzi shuyi 誘，聲色不能淫，辯者不能說，智者不能動，勇者不能恐，此
p. 143)
真人之游也。
Wenzi 5, “Dao de 智者，不以德為事；勇者，不以力為暴；仁者，不以位為惠；
道德” (Wenzi shuyi 可謂一矣。
p. 233)
Wenzi 11, “Shang 智者不妄為，勇者不妄殺...
yi 上義” (Wenzi
shuyi p. 487)
Gongsun Hong
memorial, 130
BCE (Hanshu
臣聞之，仁者愛也，義者宜也，禮者所履也，智者術之原也。
58.2616)
Yuqiu Shouwang 是以知者陷愚，勇者威怯。
memorial, ca. 124
BCE (Hanshu
64.2796)
Da Dai Liji, “Zhu 仁者莫大於愛人，知者莫大於知賢，政者莫大於官賢。
yan 主言” (SBCK /)
Da Dai Liji 49,
仁者樂道，智者利道，
“Zengzi lishi 曾子
立事” (SBCK 4/6a)
Shuiyuan 16,
夫智者不妄為，勇者不妄殺。
“Tan cong 談叢”
(SBCK 16/16b)
420
• Appendix W: Lunyu Zengzi material
Lunyu 1/4
曾子曰：吾日三省吾身：為人謀而不忠乎？與朋友交而不信乎？傳不
習乎？
Lunyu 1/9
曾子曰：慎終追遠，民德歸厚矣。
Lunyu 4/15
子曰：參乎！吾道一以貫之。曾子曰：唯。子出，門人問曰：何謂
也？曾子曰：夫子之道，忠恕而已矣。
Lunyu 8/3
曾子有疾，召門弟子曰：啟予足！啟予手！詩云：戰戰兢兢，如臨深
淵，如履薄冰。而今而後吾知免夫！小子！
Lunyu 8/4
曾子有疾，孟敬子問之。曾子言曰：鳥之將死，其鳴也哀；人之將
死，其言也善。君子所貴乎道者三：動容貌，斯遠暴慢矣；正顏色，
斯近信矣；出辭氣，斯遠鄙倍矣。籩豆之事，則有司存。
Lunyu 8/5
曾子曰：以能問於不能，以多問於寡；有若無，實若虛，犯而不校。
昔者吾友嘗從事於斯矣。
Lunyu 8/6
曾子曰：可以託六尺之孤，可以寄百里之命，臨大節而不可奪也。君
子人與？君子人也。
Lunyu 8/7
曾子曰：士不可以不弘毅，任重而道遠。仁以為己任，不亦重乎？死
而後已，不亦遠乎？
Lunyu 12/24
曾子曰：君子以文會友，以友輔仁。
Lunyu 14/26
子曰：不在其位，不謀其政。曾子曰：君子思不出其位。
Lunyu 19/16
曾子曰：堂堂乎張也，難與並為仁矣。
Lunyu 19/17
曾子曰：吾聞諸夫子：人未有自致者也，必也親喪乎！
Lunyu 19/18
曾子曰：吾聞諸夫子：孟莊子之孝也，其他可能也；其不改父之臣與
父之政，是難能也。
Lunyu 19/19
孟氏使陽膚為士師，問於曾子。曾子曰：上失其道，民散久矣。如得
其情，則哀矜而勿喜！
421
• Appendix X: Liu De’s 劉德 Hanshu biography and related passages
• Hanshu 53.2410–2411
King Xian of Hejian was established in the second year of Emperor Jing’s reign. He cultivated
learning and loved the ancient, engaging in substantive matters and pursuing the true. Whenever
he obtained a fine text from the people, he would always have a good copy made for [the donor]
and keep the original for himself, and he would solicit writings with offers of metal and silk. As
a result, men of various doctrines and techniques from the four directions thought nothing of
traveling a thousand li [to King Xian’s court]. Some had their ancestors’ old writings, so many of
which were presented to King Xian that he acquired as many books as the Han court. At this
time, [Liu] An the King of Huainan also loved writings, [but] his invited retainers mostly engaged in empty disputations. King Xian’s collected writings were all old, ancient script pre-Qin
texts like the Zhou Offices, Exalted Documents, Rituals, Ritual Records, Mengzi, and Laozi, all of
which were canons, commentaries, explanations, and records of the sort discussed by [Kongzi’s]
seventy disciples. With his learning he promoted the Six Arts and established Erudites for the
Mao Odes and the Zuo Chunqiu [=Zuozhuan]. He cultivated ritual and music, clothed himself in
the techniques of the Ru, and constantly hewed to the Ru.659 Many were the Ru of Shandong who
flocked to him.
During Emperor Wu’s reign, King Xian came to court and presented elegant music. He also
submitted his replies on the Palace of the Threefold Harmony as well as on more than thirty other matters raised in imperial edicts and queries. In his replies he advocated the Way and [appropriate] techniques, grasping the heart of the matter with a succinct and clearly expressed style.
河間獻王德以孝景前二年立，修學好古，實事求是。從民得善書，必為好寫與之，留其
真，加金帛賜以招之。繇是四方道術之人不遠千里，或有先祖舊書，多奉以奏獻王者，故
得書多，與漢朝等。是時，淮南王安亦好書，所招致率多浮辯。獻王所得書皆古文先秦舊
書，周官、尚書、禮、禮記、孟子、老子之屬，皆經傳說記，七十子之徒所論。其學舉六
藝，立毛氏詩、左氏春秋博士。修禮樂，被服儒術，造次必於儒者。山東諸儒
（者）〔多〕從而游。660
659.This line is a probable allusion to Lunyu 4/5: .
660.The Hanshu biography appears to expand upon the much shorter Shiji version (Shiji 59.2093): “King Xian of
Hejian, [Liu] De, in the second year of Emperor Jing’s reign, took advantage of his position as imperial prince to
become King of Hejian. He loved Ru learning and constantly hewed to the Ru. Many were the Ru of Shandong who
flocked to him” (河閒獻王德，以孝景帝前二年用皇子為河閒王。好儒學，被服造次必於儒者。山東諸儒多
從之游。 ).
422
武帝時，獻王來朝，獻雅樂，對三雍宮及詔策所問三十餘事。其對推道術而言，得事
之中，文約指明。
• Hanshu book 30, the “Yiwen zhi” (Hanshu 30.1708, 30.1912, 30.1726)
[The entry for the Shi:] ...There is also the learning of Master Mao, who claimed it was handed
down by Zixia. King Xian of Hejian preferred it, but it has not yet been established [as an imperially sponsored canon].
又有毛公之學，自謂子夏所傳，而河間獻王好之，未得立。
*
[The entry for the Yue:] ...At the time of Emperor Wu, King Xian of Hejian loves the Ru and together with Master Mao and others selected passages from the Zhouguan and the masters’ discussions of musical matters to make a Record of Music. He [also] presented the dance of the
“eight rows,” which was not far off from Mr. Zhi’s version...
武帝時，河間獻王好儒，與毛生等共采周官及諸子言樂事者，以作樂記，獻八佾之舞，與
制氏不相遠。
*
[The entry for Ru texts:] King Xian of Hejian’s responses to the emperor and his officials concerning the Palaces of the Three-fold Harmony, in three sections
河間獻王對上下三雍宮三篇。
423
BIBLIOGRAPHY
Primary sources
Congshu jicheng chubian 叢書集成初編. Shanghai: Shangwu yinshuguan. 1935.
Dingzhou Hanmu zhujian Lunyu 定州漢墓竹簡論語. Hebeisheng wenwu yanjiusuo 河北省文
物研究所. Beijing: Wenwu. 1997.
Guodian chumu zhujian 郭店楚墓竹簡. Jingmenshi bowuguan 荊門市博物館. Beijing: Wenwu.
1998.
Han Feizi xin jiaozhu 韓非子新校注. Compiled by Chen Qiyou 陳奇猷. Shanghai: Shanghai
guji chubanshe. 2000.
Hanshu 漢書. Compiled by Ban Gu 班固. Beijing: Zhonghua shuju. 1962.
Hou Hanshu 後漢書. Compiled by Fan Ye 范曄. Beijing: Zhonghua shuju. 1973.
Huainanzi jishi 淮南子集釋. Compiled by He Ning 何寧. Beijing: Zhonghua shu ju. 1998.
Huangdi neijing yizhu 黃帝內經譯注. By Yao Chunpeng 姚春鵬. Beijing: Zhonghua shuju.
2010.
Kaoxin lu 考信錄. By Cui Shu 崔述. Taibei: Taiwan shangwu yinshu guan. 1968.
Kongzi jiyu jiaobu 孔子集語校補. Compiled by Sun Xingyan 孫星衍 et al. Jinan: Qi Lu shushe.
1998.
Liu Zongyuan ji 柳宗元集. Beijing: Zhonghua shuju. 1979.
Lunheng jiaoshi 論衡校釋. Compiled by Huang Hui 黃暉. Beijing: Zhonghua shuju. 1990.
Lunyu jishi 論語集釋. Compiled by Cheng Shude 程樹德. Beijing: Zhonghua shuju. 1990.
Lunyu zhengyi 論語正義. In Shisan jing zhushu (fu jiaokanji) 十三經注疏 (附校勘記), edited by
Ruan Yuan 阮元. Beijing: Zhonghua shuju. 1980.
424
Lunyu zhuzi suoyin 論語逐字索引. The ICS Ancient Texts Concordance Series. Hong Kong:
Shangwu yinshuguan. 1995.
Mozi jiaozhu 墨子校注. By Wu Yujiang 吳毓江. Beijing: Zhonghua shuju. 1993.
Quan Shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文. Compiled
by Yan Kejun 嚴可均. Beijing: Zhonghua shuju. 1958.
Sanguo zhi 三國志. Compiled by Chen Shou 陳壽. Beijing: Zhonghua shuju. 1959.
Shanghai bowuguan cang Zhanguo Chu zhushu 上海博物館藏戰國楚竹書, vol. 1–8. Edited by
Ma Chengyuan 馬承源. Shanghai: Shanghai guji chubanshe. 2001.
Shiji 史記. Compiled by Sima Qian 司馬遷. Beijing: Zhonghua shuju. 1959.
Sibu congkan (chubian) 四部叢刊 (初編). Shanghai: Shangwu yinshuguan. 1922. Digitized by
the Unihan Digital Technology Co., Ltd., in 2001. Version 1.0.
Sishu zhangju jizhu 四書章句集注. Compiled by Zhu Xi 朱熹. Beijing: Zhonghua shuju. 1983.
Suishu 隨書. Compiled by Wei Zheng 魏徵. Beijing: Zhonghua shuju. 1973.
Wei Liaozi jiaozhu 尉繚子校注. By Zhong Zhaohua 鐘兆華. Zhengzhou: Zhongzhou
shuhuashe. 1982.
Wenzi shuyi 文子疏義. By Wang Liqi 王利器. Beijing: Zhonghua shuju. 2000.
Xinshu jiaozhu 新書校注. By Yan Zhenyi 閻振益 & Zhong Xia 鐘夏. Beijing: Zhonghua shuju.
2000.
Yi Zhoushu jiaobu zhuyi 逸周書校補注譯. By Huang Huaixin 黃懷信. Xi’an Shi: San Qin
chubanshe. 2006.
425
Secondary sources and translations
Ahbel-Rappe, Sara & Kamtekar, Rachana. 2006. A Companion to Socrates. Oxford: Blackwell.
Allan, Sarah. 2009. “Not the Lun Yu: The Chu Script Bamboo Slip Manuscript, Zigao, and the
Nature of Early Confucianism.” Bulletin of the School of Oriental and African Studies 72:
115-151.
Ariel, Yoav. 1996. K’ung-ts’ung-tzu: the K’ung Family Masters’ Anthology - a Study and Translation of Chapters 1-10, 12-14. Princeton: Princeton University Press.
Baxter, W. & Sagart, L. 2012. “Baxter-Sagart Old Chinese Reconstruction (Version 1.00).”
Http://crlao.ehess.fr/document.php?id=1217 (accessed 3/1/2012).
Bielenstein, Hans. 1980. The Bureaucracy of Han Times. Cambridge: Cambridge University
Press.
Bodde, Derk. 1938. China’s First Unifier; a Study of the Ch’In Dynasty as Seen in the Life of Li
Ssæu. (280?-208 B.C.). Leiden: Brill.
Brooks, E. Bruce & Brooks, A. Taeko. 1997. The Original Analects : Sayings of Confucius and
His Successors, 0479-0249. New York: Columbia University Press.
Chan, Wing-Tsit. 1963. A Source Book in Chinese Philosophy. Princeton: Princeton University
Press.
Chen Tongsheng 陳桐生. 2006. “Kongzi yulu de jieben he fanti 孔子語錄的節本和繁本.”
Kongzi yanjiu 2006: 116–122.
———. 2004. “Kongzi shilun” yanjiu 《孔子詩論》研究. Beijing: Zhonghua shuju.
Chen Xionggen 陳雄根 & He Zhihua 何志華. 2007. Xian-Qin Liang-Han dianji yin “Zhouyi”
ziliao huibian; Xian-Qin Liang-Han dianji yin “Lunyu” ziliao huibian; Xian-Qin Liang-Han
426
dianji yin “Mengzi” ziliao huibian 先秦兩漢典籍引《周易》資料彙編 ; 先秦兩漢典籍引
《論語》資料彙編 ; 先秦兩漢典籍引《孟子》資料彙編. Hong Kong: Zhongwen daxue
chubanshe.
Cook, Scott. 1995. “Yue Ji” 樂記 — Record of Music: Introduction, Translation, Notes, and
Commentary.” Asian Music 26: 1-96.
Creel, Herrlee Glessner. 1949. Confucius, the Man and the Myth. New York: J. Day Co.
Csikszentmihalyi, Mark. 2006. Readings in Han Chinese Thought. Indianapolis, IN: Hackett
Pub. Co.
———. 2004. Material Virtue : Ethics and the Body in Early China. Leiden: Brill.
———. 2002. “Confucius and the Analects in the Han.” In Confucius and the Analects: New Essays, ed. Bryan Van Norden, pp. 134-162. Oxford: Oxford University Press.
———, Mark. 2001. “Confucius.” In The Rivers of Paradise: Moses, Buddha, Confucius, Jesus,
and Muhammad as Religious Founders, eds. David Noel Freedman & Michael James Mcclymond, pp. 233–308. Cambridge: W.B. Eerdmans Pub.
Csikszentmihalyi, Mark & Nylan, Michael. 2003. “Constructing Lineages and Inventing Traditions Through Exemplary Figures in Early China.” T’oung Pao 90: 59-99.
De Crespigny, Rafe. 2007. A Biographical Dictionary of Later Han to the Three Kingdoms
(23-220 Ad). Leiden: Brill.
Denecke, Wiebke. 2010. The Dynamics of Masters Literature: Early Chinese Thought From
Confucius to Han Feizi. Cambridge: Harvard University Asia Center for the Harvard-Yenching Institute & Harvard University Press.
Dew, Nicholas. 2009. Orientalism in Louis Xiv’s France. Oxford: Oxford University Press.
Döring, Kurt. 2011. “The Students of Socrates.” In The Cambridge Companion to Socrates, ed.
427
Donald R. Morrison, pp. 24–47. Cambridge: Cambridge University Press.
Dorion, Louis-André. 2011. “The Rise and Fall of the Socratic Problem.” In The Cambridge
Companion to Socrates, ed. Donald R. Morrison, pp. 1–23. Cambridge: Cambridge University Press.
Dubs, Homer H. 1955. The History of the Former Han Dynasty, by Pan Ku. Volumes 1–3. Baltimore: Waverly Press.
Ebrey, Patricia Buckley. 1996. The Cambridge Illustrated History of China. New York: Cambridge University Press.
Eno, Robert. 1989. The Confucian Creation of Heaven: Philosophy and the Defense of Ritual
Mastery. Albany: State University of New York Press.
Fischer, Paul. 2009. “Intertextuality in Early Chinese Masters-Texts: Shared Narratives in Shi
Zi.” Asia Major 22: 1–34. :
Foucault, Michel. 1979. “What is an Author?” In Textual Strategies: Perspectives in Post-Structuralist Criticism, ed. Josué V Harari, pp. 141–160. Ithaca: Cornell University Press.
Fu Yunlung. 1981. “Studies on Confucius Since Construction (Since 1949).” Chinese Studies in
Philosophy 12: 25-51.
Fukui Shigemasa 福井重雅. 1997. “Tō Chujo No Taisaku No Kisoteki Kenkyū 董仲舒の対策
の基礎的研究.” Shigaku zasshi 106: 157–204.
———. 1995. “Shin Kan Jidai Ni Okeru Hakase Seido No Tenkai: Gokei Hakase No Secchi O
Meguru Gigi Sairon 秦漢代時代における博士制度の再論展開：五経博士の設置をめぐ
る疑義再論.” Tōyōshi kenkyū 54: 1–31.
———. 1994. “Rikukei, Rikugei to Gokei: Kandai Ni Okeru Gokei No Seiritsu 六経六芸と五
経：漢代における五経の成立.” Chūgoku shigaku 4: 139–164.
428
Goldin, Paul R. 2011. “Confucius and His Disciples in the Analects: Or, Why the Traditional
Chronology is Right.” A paper presented at “The Analects: A Western Han Text?”, a conference held at Princeton University, November 4–5, 2011.
———. “When Zhong 忠 Does Not Mean “Loyalty.” Dao 7: 165–174.
———. “The Thirteen Classics.” In The Columbia History of Chinese Literature, ed. Victor H
Mair, pp. xx, 1342 p. New York: Columbia University Press.
Goodman, Howard L. 1998. Ts’ao P’i Transcendent : The Political Culture of Dynasty-Founding in China At the End of the Han. Seattle & Surrey, England: Scripta Serica (Distributed
Worldwide by Curzon Press, Ltd.).
Graham, A. C. 1989. Disputers of the Tao : Philosophical Argument in Ancient China. La Salle,
Ill.: Open Court.
Graham, A.C. 1972. “The Classical Chinese Topic-Marker “Fu” 夫.” Bulletin of the School of
Oriental and African Studies, University of London 35: 85–110.
Gu Jiegang 顧頡剛. 1982 (rpt). “Chunqiu shi de Kongzi he Han dai de Kongzi 春秋時的孔子和
漢代的孔子,” in Gushi bian 古史辨 vol. 2, pp. 130–139. Shanghai: Shanghai guji
chubanshe.
Guojia wenwuju guwenxian yanjiushi 國家文物居古文獻研究室 et al. 1981. “‘Rujia zhi yan’
shiwen”《儒家之言》釋文. Wenwu 8: 13-19
Han Ziqiang 韓自強. 2004. Fuyang Hanjian “Zhouyi” yanjiu, fu “Rujia zhe yan” zhangti,
“Chunqiu shiyu” ji xiangguan zhujian 阜陽漢簡《周易》研究，附《儒家者言》章題、
《春秋事語》章題及相關竹簡. Shanghai: Shanghai guji chubanshe.
Harbsmeier, Christoph. 1990. “Confucius Ridens: Humor in the Analects.” Harvard Journal of
Asiatic Studies 50.1: 131–161.
429
Hawkes, David. 1985. The Songs of the South: An Ancient Chinese Anthology of Poems By Qu
Yuan and Others. New York: Penguin Books.
Henderson, John B. 1991. Scripture, Canon, and Commentary: A Comparison of Confucian and
Western Exegesis. Princeton: Princeton University Press.
Henry, Eric. 1999. “‘Junzi Yue’ Versus ‘Zhongni Yue’ in Zuozhuan.” Harvard Journal of Asiatic Studies 59: 125-161.
Hightower, James Robert. 1952. Han Shih Wai Chuan: Han Ying’s Illustrations of the Didactic
Application of the Classic of Songs. Cambridge: Harvard University Press.
Hu Pingsheng 胡平生 & Zhang Defang 張德芳. 2001. Dunhuang Xuanquan Han jian shicui 敦
煌懸泉漢簡釋粹. Shanghai: Shanghai guji chubanshe.
Huang Huaixin 黃懷信. 2004. Shanghai bowuguan cang Zhanguo Chu zhushu “Shilun” jieyi 上
海博物館藏戰國楚竹書《詩論》解義. Beijing: Shehui kexue wenxian chubanshe.
Huang Jinxing 黃進興. 1998. You ru shengyu: quanli, xinyang yu zhengdangxing 優入聖域：權
力、信仰與正當性. Xian: Shaanxi Shifan daxue chubanshe.
Hunter, Michael & Kern, Martin. Forthcoming. “Quotation.” In Reading Early Chinese Manuscripts: Texts, Contexts, Methods, ed. Martin Kern, Dirk Meyer, and Wolfgang Behr. Leiden:
Brill.
Hunter, Michael, Martin Kern & Oliver Weingarten (eds.). Forthcoming. Untitled Lunyu project.
Ivanhoe, P. J. & Van, Norden, Bryan W. 2005. Readings in Classical Chinese Philosophy. Indianapolis: Hackett Pub.
Guanghui, Jiang. 2008. “A Modern Translation of Confucius’s Comments on the Poetry (Kongzi
Shilun).” Contemporary Chinese Thought 39: 49-60.
Hu Zhikui 胡志奎. 1978. Lunyu bianzheng 論語辨證. Taibei: Lianjing chuban shiye gongsi.
430
Itō Jinsai 伊藤仁齋. 2007. “Rongo kogi: Sōron 論語古義—総論.” In Itō Jinsai shū 伊藤仁齋集,
ed. Kimura Eiichi 木村英一, pp. 315-330. Tōkyō: Chikuma shobō.
Jiang Yihua 姜義華, Zhang Ronghua 張榮華, & Wu Genliang 吳根梁. 1990. Kongzi—Zhou Qin
Han Jin wenxianji 孔子—周秦漢晉文獻集. Shanghai: Fudan daxue chubanshe.
Kaneto Mamoru 金戶守. 1981. “Shiki “Rongo” kō, yozoku: Kansho yori mita Rongo seiritsu kō
史記「論語」考、余続：漢書より見た論語成立考.” Shitennōji Joshi Daigaku Kiyō 13:
17–29.
———. 1980. “Shiki “Rongo” kōyo 史記「論語」考余.” Shitennōji Joshi Daigaku Kiyō 12:
21–32.
———. 1979. “Shiki “Rongo” kō, sono roku: Sōketsu no seiritsu ni tsuite 史記「論語」考、そ
の六：總結の成立について.” Shitennōji Joshi Daigaku Kiyō 11: 19–26.
———. 1978. “Shiki “Rongo” kō, sono go: Hakui retsuden o ronjite Rongo Kōshi gen ni oyobu
史記「論語」考、その五：伯夷列伝を論じて論語孔子言に及ぶ.” Shitennōji Joshi
Daigaku Kiyō 10: 65–79.
———. 1977. “Shiki “Rongo” kō, sono yon: Seika inyō Kōshi yori Rongo no seiritsu ni tsuite
史記「論語」考、その四：世家引用孔子言より論語の成立について.” Shitennōji Joshi
Daigaku Kiyō 9: 101–118.
———. 1976. “Shiki “Rongo” kō, sono san: Hongi • Hassho no Rongo seiritsu shiryō ni tsuite
史記「論語」考、その三：本紀 • 八書の論語成立資料について.” Shitennōji Joshi
Daigaku Kiyō 8: 45–59.
———. 1975. “Shiki “Rongo” kō, sono ni: Chūji deshi retsuden yori Rongo seiritsu ni tsuite 史
記「論語」考、その二：仲尼弟子列伝より論語成立について.” Shitennōji Joshi
Daigaku Kiyō 7: 51–70.
431
———. 1974. “Shiki ‘Rongo’ kō: Kōshi seika, sono “Rongo” seiritsu shiron 史記「論語」考：
世家、その「論語」成立試論.” Shitennōji Joshi Daigaku Kiyō 6: 14–38.
———. 1973. “Kobun Rongo kō: Shiki deshi iu, Kansho Geibunshi 古文論語考：史記弟子
云，漢書藝文志.” Shitennōji Joshi Daigaku Kiyō 5: 43–59.
———. 1972. “Kōshi to sono jidai: Rongo hihan oboegaki 孔子とその時代：論語批判覚書.”
Shitennōji Joshi Daigaku Kiyō 4: 101–114.
Kano Naoki 狩野直喜 (1868-1947). 1977. Rongo Moshi kenkyu 論語孟子研究. Tokyo: Misuzu.
Kennedy, George Alexander. 1991. On Rhetoric: A Theory of Civic Discourse. New York:
Oxford University Press.
Kern, Martin. 2011. “Confucius as Author and Critic.” A paper presented at “The Analects: A
Western Han Text?”, a conference held at Princeton University, November 4–5, 2011.
———. “Lost in Tradition: The Classic of Poetry We Did Not Know.” Hsiang Lectures on Chinese Poetry 5: 29–56.
———. 2007. “Beyond the Mao Odes: Shijing Reception in Early Medieval China.” Journal of
the American Oriental Society 127: 131–142.
———. 2005. “Quotation and the Confucian Canon in Early Chinese Manuscripts: The Case of
“Zi Yi” (Black Robes).” Asiatische Studen/Études Asiatiques 59: 293-332.
———. 2003. “Early Chinese Poetics in the Light of Recently Excavated Manuscripts.” In Recarving the Dragon: Understanding Chinese Poetics, ed. Olga Lomová, pp. 27-72. Prague:
Karolinum.
———. 2001. “Ritual, Text, and the Formation of the Canon: Historical Transitions of Wen in
Early China.” T’oung Pao 87: 43-91.
———. 2000. The Stele Inscriptions of Ch’in Shih-Huang: Text and Ritual in Early Chinese Im432
perial Representation. New Haven: American Oriental Society.
Kimura Eiichi 木村英一. 1971. Kōshi to Rongo 孔子と論語. Tōkyō: Sōbunsha.
Klein, Esther. 2010. The History of a Historian: Perspectives on the Authorial Roles of Sima
Qian. Ph.D. thesis, Princeton University.
Knechtges, David R. 1987. Wen Xuan, Or, Selections of Refined Literature, Volume 2. Princeton:
Princeton University Press.
Knoblock, John & Riegel, Jeffrey. 2000. The Annals of Lü Buwei : A Complete Translation and
Study. Stanford: Stanford University Press.
Kunio Seo. 1992. Kôshi, Môshi Ni Kansuru Bunken Mokuroku 孔子、孟子に関する文献目録.
Tōkyō: Hakuteisha.
Lang, Mabel L. 1978. Socrates in the Agora. Princeton: American School of Classical Studies at
Athens.
Lau, D.C. 2000. The Analects - 2000 Bilingual Ed. Hong Kong: Chinese University Press.
———. 1970. Mencius. New York: Penguin Books.
Le Blanc, Charles. 1985. Huai-Nan Tzu: Philosophical Synthesis in Early Han Thought: The
Idea of Resonance (Kan-Ying) With a Translation and Analysis of Chapter Six. Hong Kong:
Hong Kong University Press.
Lee Songsi 李成市, Yun Yonggu 尹龍九, & 金慶浩 Kim Kyŏng-ho. 2011. “Pingrang Zhenbodong 364 Hao Mu Chutu Zhujian Lunyu 平壤貞柏洞 364 號墓出土竹簡《論語》.”
Chutu wenxian yanjiu 10: 158–194.
———. 2010. “平壤貞柏洞三六亖號墓出土竹簡“論語”につぃて.” Chūgoku shutsudo shiryō
kenkyū 14: 110–149.
———. 2009. “P’yŏngyang Chŏng-Baek-Dong 364 平壤貞柏洞 364 號墳 출토 竹簡『論語』
433
에 대하여.” 木簡與文字 4: 127–166.
Legge, James. 1991. The Chinese Classics, Vol. 5: The Ch’Un Ts’Ew With the Tso Chuen.
Taibei: SMC Publishing.
———. The Life and Teachings of Confucius, With Explanatory Notes. Philadelphia: J.B. Lippincott & Co.
Levi, Jean. 1995. “Quelques Exemples De Détournement Subversif De La Citation Dans La Littérature Classique Chinoise.” Extrême-Orient, Extrême-Occident 17: 41–65.
Lewis, Mark Edward. 1999. Writing and Authority in Early China. Albany: State University of
New York Press.
Li Ling 李零. 2004. Jianbo gushu yu xueshu yuanliu 簡帛古書與學術源流. Beijing: Sanlian.
Li Qiqian 李
謙 et al. (ed.). 1991a. Kongzi dizi ziliao huibian 孔子弟子資料彙編. Jinan: Shan-
dong youyi shushe.
———. 1991b. Kongzi ziliao huibian 孔子資料彙編. Jinan: Shandong youyi shushe.
Li Xueqin 李學勤. 2004. “Shangbo Chu jian “Lu bang da han” jieyi 上博楚簡《魯邦大旱》解
義.” Kongzi yanjiu 1.
———. 2000. “The Confucian Texts From Guodian Tomb Number One: Their Date and Significance.” In The Guodian Laozi: Proceedings of the International Conference, Dartmouth College, May 1998, eds. Sarah Allan & Crispin Williams, pp. 107-111.
Liao Mingchun 廖名春. 2004. “Shangbo cang Chu jian “Lu bang da han” jiaobu 上博藏楚簡
《魯邦大旱》校補.” Guji zhengli yanjiu xuekan 1: 4–9.
Loewe, Michael. 2011. Dong Zhongshu, a “Confucian” Heritage and the Chunqiu Fanlu. Leiden ; Boston: Brill.
———. 2004. The Men Who Governed Han China : Companion to a Biographical Dictionary of
434
the Qin, Former Han and Xin Periods. Leiden: Brill.
———. 2000. A Biographical Dictionary of the Qin, Former Han and Xin Periods (221 Bc - Ad
24). Leiden: Brill.
———. 1993. Early Chinese Texts : A Bibliographical Guide. University of California, Berkley:
Society for the study of early China and the Institute of East Asian Studies.
———. 1986. “The Former Han Dynasty.” In The Cambridge History of China, Volume I: The
Ch’in and Han Empires, 221 B.C.—A.D. 220, eds. Denis Crispin Twitchett & John King
Fairbank, pp. 193–222. Cambridge: Cambridge University Press.
Luo Anxian 羅安憲. 2008. Zhongguo Kongxue shi 中國孔學史. Beijing: Renmin chubanshe.
Lynn, Richard John. 1994. The Classic of Changes: A New Translation of the I Ching as Interpreted By Wang Bi. New York: Columbia University Press.
Ma Guohan 馬國翰. 1990. Yuhan shanfang jiyi shu 玉函山房輯佚書. Shanghai: Shanghai guji
chubanshe.
Maeder, Erik W. 1992. “Some Observations on the Composition of the “Core Chapters” of the
Mozi.” Early China 17: 27-82.
Major, John S. et al. 2010. The Huainanzi : A Guide to the Theory and Practice of Government
in Early Han China. New York: Columbia University Press.
Makeham, John. 2011. “A Critical Overview of Some Contemporary Chinese Perspectives on
the Composition and Date of Lunyu.” A paper presented at “The Analects: A Western Han
Text?”, a conference held at Princeton University, November 4–5, 2011.
———. 2009. “In Quest of the Authentic Confucius.” China Heritage Quarterly 19.
———. 1999. “He Yan, Xuanxue and the Editorship of the Lunyu Jijie.” Early Medieval China
5: 1-35.
435
———. 1998. “Between Chen and Cai: Zhuangzi and the Analects.” In Wandering At Ease in
the Zhuangzi, ed. Roger T. Ames, pp. 75–100. Albany: State University of New York Press.
———. 1997. “The Earliest Extant Commentary on Lunyu: Lunyu Zheng Shi zhu.” T’oung Pao
83: 260-299.
———. 1996. “The Formation of Lunyu as a Book.” Monumenta Serica 44: 1-24.
McClymond, Michael James. 2001. “Jesus.” In The Rivers of Paradise: Moses, Buddha, Confucius, Jesus, and Muhammad as Religious Founders, eds. David Noel Freedman & Michael
James Mcclymond, pp. 309–456. Cambridge: W.B. Eerdmans Pub.
Momigliano, Arnaldo. 1993. The Development of Greek Biography. Cambridge: Harvard University Press.
Moore, Cameron. 2011. “Rongo, the State of the Field: A Review of Japanese Scholarship on the
Composition and Transmission of the Lunyu in the Han.” A paper presented at “The
Analects: A Western Han Text?”, a conference held at Princeton University, November 4–5,
2011.
Nienhauser, William H. 1995. The Grand Scribe’s Records, Vol. 7: The Memoirs of Pre-Han
China. Bloomington: Indiana University Press.
Nivison, David Shepherd. 1999. “The Classical Philosophical Writings.” In The Cambridge History of Ancient China : From the Origins of Civilization to 221 B.C, ed. Michael Loewe &
Edward L Shaughnessy, pp. 745-812. Cambridge: Cambridge University Press.
Nylan, Michael. 1999. “A Problematic Model: The Han “Orthodox Synthesis,” Then and Now.”
In Imagining Boundaries: Changing Confucian Doctrines, Texts, and Hermeneutics, eds.
Kai-Wing ChowOn-Cho Ng & John B. Henderson, pp. 17-56. Albany: SUNY Press.
———. 1994. “The Chin Wen/Ku Wen Controversy in Han Times.” T’oung Pao 80: 83-145.
436
Nylan, Michael & Wilson, Thomas A. 2010. Lives of Confucius: Civilization’s Greatest Sage
Through the Ages. New York: Doubleday.
Owen, Stephen. 2001. “Liu Xie and the Discourse Machine.” In A Chinese Literary Mind: Culture, Creativity, and Rhetoric in Wenxin Diaolong, ed. Zong-qi Cai, pp. 175–192. Stanford:
Stanford University Press.
———. 1992. Readings in Chinese Literary Thought. Cambridge: Council on East Asian Studies, Harvard University (Distributed by Harvard University Press).
Paget, James Carleton. 2001. “Quests for the Historical Jesus.” In The Cambridge Companion to
Jesus, ed. Markus Bockmuehl, pp. 138–155. Cambridge: Cambridge University Press.
Pelliot, Paul. 1916. “Le Chou King En Caractères Anciens Et Le Chang Chou Che Wen.” In Mémoires Concernant L’Asie Orientale (Inde, Asie Centrale, Extrême-Orient), Paris: E. Leroux.
Pines, Yuri. 2009. Envisioning Eternal Empire: Chinese Political Thought of the Warring States
Era. Honolulu: University of Hawai°i Press.
Regier, Willis Goth. 2010. Quotology. Lincoln: University of Nebraska Press.
Richter, Matthias. 2011. “Do Early Chinese Book Formats Explain Textual Structures in the
Analects?” A paper presented at “The Analects: A Western Han Text?”, a conference held at
Princeton University, November 4–5, 2011.
Richter, Matthias. 2005. Guan Ren : Texte Der Altchinesischen Literatur Zur Charakterkunde
Und Beamtenrekrutierung. Bern: Lang.
Riegel, Jeffrey. 1986. “Poetry and the Legend of Confucius’ Exile.” Journal of the American
Oriental Society 106: 13-22.
———. 1978. “The Four ‘Tzu Ssu’ Chapters of the ‘Li chi’: An Analysis and Translation of the
‘Fang Chi,’ ‘Chung Yung,’ ‘Piao Chi,’ and ‘Tzu I.’. Ph.D. thesis, Stanford University.
437
Schaberg, David. 2005. “Platitude and Persona: Junzi Comments in the Zuozhuan and Beyond.”
In Historical Truth, Historical Criticism, and Ideology: Chinese Historiography and Historical Culture From a New Comparative Perspective, ed. Helwig Et al. Schmidt-Glintzer,
———. 2001. A Patterned Past: Form and Thought in Early Chinese Historiography. Cambridge: Harvard University Asia Center (Distributed by Harvard University Press).
———. 2001. “Review: “Sell it! Sell it!”: Recent Translations of Lunyu.” Chinese Literature:
Essays, Articles, Reviews (CLEAR) 23: 115-139.
Schilling, Dennis. 2006. “Imitatio Confucii: Eine Studie Zur Poetologie Der Nachahmung in
Yang Xiong’s Fayan.” In Han-Zeit : Festschrift Für Hans Stumpfeldt Aus Anlass Seines 65.
Geburtstages, eds. Hans StumpfeldtMichael FriedrichHans Van Ess & Reinhard Emmerich,
pp. 397-421. Wiesbaden: Harrassowitz.
Schultz, Richard L. 1999. The Search for Quotation: Verbal Parallels in the Prophets. Sheffield,
England: Sheffield Academic Press.
Schwartz, Benjamin Isadore. 1985. The World of Thought in Ancient China. Cambridge: Belknap
Press of Harvard University Press.
Shaheen, Joel. 2002. “An Annotated Bibliography of Works on Confucius and the Analects.” In
Confucius and the Analects: New Essays, ed. Bryan Van Norden, pp. 303-320.
Shryock, John Knight. 1966. The Origin and Development of the State Cult of Confucius. New
York: Paragon Book Reprint Corp.
Simson, Wojciech Jan. 2011. “Did Mengzi Teach the Analects?” A paper presented at “The
Analects: A Western Han Text?”, a conference held at Princeton University, November 4–5,
2011.
———. 2006. Die Geschichte Der Aussprüche Des Konfuzius (Lunyu). Bern ; New York: P.
438
Lang.
Slingerland, Edward. 2003. Confucius’ Analects, With Selections From Traditional Commentaries. Indianapolis: Hackett Publishing Company.
———. 2000. “Why Philosophy is Not “Extra” in Understanding the Analects, a Review of the
Original Analects, By E. Bruce Brooks & a. Taeko Brooks.” Philosophy East and West 50:
137-141.
Takeuchi Yoshio 武内義雄. 1972. Rongo No Kenkyū 論語の研究. Tōkyō: Iwanami Shoten.
Tsuda Sōkichi 津田左右吉. 1946. Rongo to Kōshi no shisō 論語と孔子の思想. Tokyo:
Iwanami.
Tuckett, Christopher. 2001. “Sources and Methods.” In The Cambridge Companion to Jesus, ed.
Markus Bockmuehl, pp. 122–137. Cambridge: Cambridge University Press.
Van Els, Paul. “Confucius’s Sayings Entombed: On Han Dynasty Analects Manuscripts.” A paper presented at “The Analects: A Western Han Text?”, a conference held at Princeton University, November 4–5, 2011.
Van Norden, Bryan. 2002. “Unweaving the “One Thread” of Analects 4:15.” In Confucius and
the Analects: New Essays, ed. Bryan Van Norden, pp. 216–236. Oxford ; New York: Oxford
University Press.
Vander, Waerdt, Paul A. 1994. The Socratic Movement. Ithaca: Cornell University Press.
Vankeerberghen, Griet. 2001. The Huainanzi and Liu An’s Claim to Moral Authority. Albany:
State University of New York Press.
Waley, Arthur. 1938. The Analects of Confucius. London: G. Allen & Unwin ltd.
Waley, Arthur. 1996. The Book of Songs. New York: Grove Press.
Wang Baoxuan 王葆玹. 2000. “A Discussion of the Composition Dates of the Various Guodian
439
Chu Slip Texts and Their Background.” Contemporary Chinese Thought 32: 18–42.
Watsuji Tetsurō. 1948. Kōshi. Tōkyō: Uemura Shoten.
Weingarten, Oliver. 2010. Textual Representations of a Sage: Studies of Pre-Qin and Western
Han sources on Confucius (551–479 BC). Ph.D. thesis, Cambridge University.
———. 2009. “Confucius and Pregnant Women: An Investigation Into the Intertextuality of the
Lunyu.” Journal of the Americal Oriental Society 129: 1–22.
Wu Zhenxun 伍振勳. 2009. “”Shengren Xushi Yu Shensheng Dianfan 聖人敘事與神聖典範”.”
Qinghua xuebao 39: 227–259.
Yang, Xianyi & Yang, Gladys. 1979. Selections From Records of the Historian. Peking: Foreign
Languages Press.
Zhang Zuwei 張祖偉. 2008. “Cong Shiji, Hanshu, Hou Hanshu Kan Lunyu Zai Liang Han De
Liuchuan 從《史記》、《漢書》、《後漢書》看《論語》在兩漢的流傳.” M.A. thesis,
Shandong University.
Zhao Zhenxin 趙貞信, ed. 1996. Lunyu Bian 論語辨. Shanghai: Shanghai shudian.
———. 1962. “Lunyu “Yao Yue” Zhang Tanyuan 論語《堯曰》章探源.” Beijing shifan daxue
xuebao.
———. 1961. “”Lunyu” jiujing shi shei bianzuan de《論語》究竟是誰編纂的.” Beijing Shifan
Daxue xuebao 4: 11–24.
———. 1936. “論語一名之來歷及其解釋.” Shixue jikan 1: Beiping yanjiu yuan.
Zhongguo shehui kexue zhexue yanjiusuo 科學院哲學研究所. 1987. Kongzi yanjiu lunwen
zhuzuo mulu 孔子研究論文著作目錄, 1949-1986. Jinan: Qi Lu shushe. 1987.
Zhu Weizheng 朱維錚. 2002. Zhongguo jingxue shi shi jiang 中國經學史十講. Shanghai: Fu440
dan daxue chubanshe.
———. 1987. “Lishi de Kongzi he Kongzi de lishi 歷史的孔子和孔子的歷史.” In Kongizi yanjiu lunwen ji 孔子研究論文集, ed. Zhonghua Kongzi yanjiusuo 中華孔子研究所, pp. 156–
172. Beijing: Jiaoyu kexue chubanshe.
———. 1986. “Lunyu jieji cuoshuo 論語結集脞說.” Kongzi yanjiu 孔子研究 1: 40-52.
441